DEVELOPMENT OF MICROFLUIDIC MODULES FOR DNA

PURIFICATION VIA PHENOL EXTRACTION AND ANALYTE

CONCENTRATION USING TRANSVERSE ELECTROKINETICS

by

MERCEDES C. MORALES

A dissertation submitted to the

Graduate School-New Brunswick

Rutgers, The State University of New Jersey

and

The Graduate School of Biomedical Sciences

University of Medicine and Dentistry of New Jersey

In partial fulfillment of the requirements

For the degree of

Doctor of Philosophy

Graduate Program in Biomedical Engineering

Written under the direction of

Professor Jeffrey D. Zahn

And approved by

________________________

________________________

________________________

________________________

New Brunswick, New Jersey

January, 2012

ii

ABSTRACT OF THE DISSERTATION

DEVELOPMENT OF MICROFLUIDIC MODULES FOR DNA PURIFICATION VIA

PHENOL EXTRACTION AND ANALYTE CONCENTRATION USING

TRANSVERSE ELECTROKINETICS

By MERCEDES C. MORALES

Dissertation Director: Professor Jeffrey D. Zahn

In this work, microfluidic platforms have been designed and evaluated to demonstrate

microscale DNA purification via organic (phenol) extraction as well as analyte trapping

and concentration using a transverse electrokinetic force balance.

First, in order to evaluate DNA purification via phenol extraction in a microdevice, an

aqueous phase containing protein and DNA and an immiscible receiving organic phase

were utilized to evaluate microfluidic DNA extraction under both stratified and droplet-

based flow conditions using a serpentine microfluidic device. The droplet based flow

resulted in a significant improvement of protein partitioning from the aqueous phase due

to the flow recirculation inside each droplet improving material convective transport into

the organic phase. The plasmid recovery from bacterial lysates using droplet-based flow

was high (>92%) and comparable to the recovery achieved using commercial DNA

purification kits and standard macroscale phenol extraction.

Second, a converging Y-inlet microfluidic channel with integrated coplanar electrodes

was used to investigate transverse DNA and protein migration under uniform direct

current (DC) electric fields. Negatively charged samples diluted in low and high ionic

iii

strength buffers were co-infused with a receiving buffer of the same ionic strength into a

main channel where transverse electric fields were applied. Experimental results

demonstrated that charged analytes could traverse the channel width and accumulate at

the positive bias electrode in a low electroosmotic mobility and high electrophoretic

mobility condition (high ionic strength buffer) or migrated towards an equilibrium

position within the channel when both electroosmotic mobility and electrophoretic

mobility are high (low ionic strength buffer). The different behaviors are the result of a

balance between the electrophoretic force and a drag force induced by a recirculating

electroosmotic flow generated across the channel width due to the bounding walls.

The miniaturization of DNA phenol extraction and the novel electrokinetic trapping

techniques presented in this research are the initial steps towards an efficient DNA

sample preparation chip which could be integrated with post-extraction DNA

manipulations for genomic analysis modules such as capillary electrophoretic

separations.

iv

ACKNOWLEDGEMENTS

First at all, I would like to thank my advisor, Professor Jeffrey Zahn. He gave me the

opportunity to get involved in stimulating laboratory experiences that made me acquire

technical skills as well as inspire my creativity. His mentorship and continuous

encouragement have helped me to stay motivated and guided me during the process of

this research.

In the same way, I want to thank Professor Hao Lin for his guidance as a member of my

committee, and Professor Li Cai and Professor Laura Fabris for providing helpful

comments and advises to this research.

I thank all my colleagues at the Biomedical Engineering Department for making my work

in the lab more enjoyable. I would like to specially acknowledge Larry Sasso, Kiana

Aran, Alex Fok, Jack Zheng and Jean Lo, for their support, help and sense of humor.

In addition, I would like to express my sincere appreciation to the faculty and

administrative staff at the Biomedical Engineering Department for their advice and help

during these years.

Thanks to all my friends in Argentina, in the United States and now in other countries for

all their support during this time. Thank you Nuria, Nachi, Valeria, Greg, Jose, Loreto,

Bahar, Chris, Marcos, Rocio, Alberto and Mosteiro family.

A gigantic thank-you goes to Mete Rodoper for all his care, enormous patience and

companionship during these years.

Finally, I would like to acknowledge and give my profound gratitude to my parents

Eduardo and Consuelo and my siblings Eduardo, Cecilia and Julieta for their support and

encouragement from a distance that kept me standing in the darkest times.

v

DEDICATION

A mis dos gigantes, mis padres, Eduardo y Consuelo

por todo su amor, educación y apoyo.

To my two giants, my parents, Eduardo and Consuelo

for all their love, education and support.

vi

TABLE OF CONTENTS

Abstract ............................................................................................................................... ii

Acknowledgements ............................................................................................................ iv

Dedication .......................................................................................................................... v

Table of Contents .............................................................................................................. vi

List of Tables ..................................................................................................................... x

List of Illustrations ............................................................................................................. xi

Chapter 1 ............................................................................................................................. 1

Introduction ......................................................................................................................... 1

1.1 Significance of DNA Purification .........................................................................1

1.2 The Need for Sample Pre-concentration ...............................................................4

1.3 Dissertation Overview ..........................................................................................6

Chapter 2 ............................................................................................................................. 8

Background of DNA Liquid Extraction .............................................................................. 8

2.1 Standard DNA extraction methods .......................................................................8

Phenol Extraction ..................................................................................................8

Solid Phase Extraction (SPE)..............................................................................10

2.2 Microfluidic DNA Extraction .............................................................................11

Silicon pillars ......................................................................................................11

Silica bead/sol-gel systems .................................................................................12

Other novel phases ..............................................................................................15

Microfluidic Phenol Extraction ...........................................................................16

2.3 Theoretical background: Mechanism of Partitioning .........................................17

vii

2.4 Theoretical background: Two-Phase Microfluidic Systems ...............................18

2.5 Convective Transport Enhancements .................................................................22

Mixing enhancement using droplets ...................................................................23

Chaotic Advection ..............................................................................................24

Chapter 3 ........................................................................................................................... 26

Microfluidic DNA purification via phenol extraction ...................................................... 26

3.1 Device Design .....................................................................................................26

3.2 Device fabrication ...............................................................................................29

3.3 Materials and Experiment setup..........................................................................31

Aqueous phase preparation .................................................................................31

Bacterial lysates preparation ...............................................................................32

Experimental procedure ......................................................................................33

3.4 Results and Discussion .......................................................................................34

Review of serpentine device comparison ...........................................................34

Review of flow pattern comparison ....................................................................37

Plasmid DNA Purification from Bacterial Lysates .............................................41

Chapter 4 ........................................................................................................................... 47

Background of Electrokinetic Sample Concentration....................................................... 47

4.1 Electroosmosis ....................................................................................................48

4.2 Electrophoresis ....................................................................................................50

4.3 Combined electrophoresis and electroosmosis in confining geometries ............52

4.4 Dielectrophoresis ................................................................................................56

4.5 State-of-the-art of sample concentration devices by electrokinetics forces ........58

viii

Chapter 5 ........................................................................................................................... 64

Sample Trapping and Concentration ................................................................................ 64

5.1 Device Fabrication ..............................................................................................64

5.2 Materials and Buffer Characterization ................................................................67

5.3 Experimental Setup .............................................................................................69

5.4 Results and Discussion .......................................................................................70

Transverse migration of DNA in 1× TBE ..........................................................70

Transverse migration of DNA in 1× TE .............................................................73

Transverse migration of proteins ........................................................................77

Further investigation of migration dynamics using pure AC field and pulsed

field .....................................................................................................................78

Effect of EO and EP mobility on DNA migration and concentration ................81

EO Flow Simulation ...........................................................................................84

Particle Visualization ..........................................................................................94

Chapter 6 ........................................................................................................................... 97

Conclusions and Future Directions ................................................................................... 97

6.1 Microfluidic DNA purification via phenol extraction ........................................97

Conclusions .........................................................................................................97

Future Work ........................................................................................................98

6.2 DNA trapping by electrokinetic forces .............................................................100

Conclusions .......................................................................................................100

Future work .......................................................................................................101

Bibliography ................................................................................................................... 103

ix

Appendix A ..................................................................................................................... 110

Microfluidic Channel and Electrode Fabrication ............................................................ 110

Appendix B ..................................................................................................................... 115

Sample Preparation and Measurements .......................................................................... 115

Appendix C ..................................................................................................................... 120

COMSOL Simulation Parameters................................................................................... 120

Curriculum Vitae ............................................................................................................ 122

x

LIST OF TABLES

Table 5.1. Comparison of electroosmotic mobilities for buffer used ............................... 68

Table 5.2. Zeta Potentials and EO mobilities at different surfaces for each buffer used.. 87

xi

LIST OF ILLUSTRATIONS

Figure 2.1: (a) SEM photograph of a pillar-based DNA extraction chip. (b) Integrated

microfluidic cartridge containing the pillar-based DNA chip and a PCR reaction tube [8,

9]. ...................................................................................................................................... 12

Figure 2.2: Silica particle-based microchip with sol-gel immobilization. A) 1x

magnification; B) 10x magnification; C) 500x magnification of cross section [10]. ....... 13

Figure 2.3: (a)-(e) Schematic steps of bacteria lysis and DNA purification process. (f)

Image of the DNA microfluidic chip with parallel architecture [11]. .............................. 14

Figure 2.4: Photograph of aqueous droplet formation containing cell lysates, including

green fluorescent protein GFP which partition to organic phase [20]. ............................. 16

Figure 2.5: Illustration of multiple reagent mixing in a droplet by recirculating flows [30].

........................................................................................................................................... 24

Figure 2.6: Droplets moving through a winding microchannel, (a) experimental results

and (b) schematic of the droplet folding [50]. .................................................................. 25

Figure 3.1: Fabrication steps of photolithography and soft lithography techniques. ........ 29

Figure 3.2: (a) Serpentine mask designs and (b) photograph of the microfluidic chip. ... 30

Figure 3.3: Comparison of the serpentine device design based on the protein depletion.

(a) using epi-fluorescent image analysis and (b) absorption spectra analysis. ................ 35

Figure 3.4: Serpentine device: (a) images of BSA partition into organic phase using

stratified flow, (b) the device geometry and dimensions, and (c) images of BSA partition

using droplet-based flow. .................................................................................................. 38

xii

Figure 3.5: Analysis of inlet and outlet aliquots for stratified flow using (a) dye

absorption spectra and (b) DNA gel electrophoresis (1: inlet, 2: aqueous outlet, 3: organic

outlet). ............................................................................................................................... 39

Figure 3.6: Analysis of inlet and outlet aliquots for droplet-based flow using (a) dye

absorption spectra and (b) DNA gel electrophoresis (1: inlet, 2: aqueous outlet, 3: organic

outlet). ............................................................................................................................... 40

Figure 3.7: (a) Redesigned device geometry and diamensions, (b) photograph of the new

shorter device and (c) images of rhodamine-labeled BSA partitioning from aqueous phase

into organic phase. ............................................................................................................ 42

Figure 3.8: Comparison using the redesigned microfluidic device, Qiagen kit and

standard macroscale phenol extraction: (a) gel electrophoresis analysis and (b) total

plasmid recovery from bacterial lysates at the aqueous outlet. ........................................ 43

Figure 4.1: Schematic of the EP particle motion and the EO flow in an open channel. ... 48

Figure 4.2: Net particle velocity Up as a result of the EP force and the EO flow drag

force. ................................................................................................................................. 53

Figure 4.3: EP velocity, transverse EO flow velocity and net particle velocity in a closed

glass/PDMS microchannel (cross-sectional view). (a) EO flow only (μEP=0), (b) EO flow

and high EP force (μEP>μEO), and (c) EO flow and EP force with similar mobility values

(μEP≈μEO). .......................................................................................................................... 55

Figure 4.4: FASS stacking: (a) schematic of the low-conductivity sample loading into a

high-conductivity buffer and (b) epi-fluorescence images during the stacking [82]. ....... 59

Figure 4.5: Images of isotachophoresis stacking of two samples with different

electrophoretic mobility [88]. ........................................................................................... 60

xiii

Figure 4.7. Trapping of DNA molecules at the edges of gold-film strips using

dielectrophoresis. (a) and (c) before field application, (b) and (d) 1 min after applying a

30-Hz, 200-V p-p signal, nearly all of the molecules became trapped [93]. .................... 62

Figure 4.8: Chip layout and image of a photopolymerized size exclusion membrane

where analytes are trapped and concentrated [98]. ........................................................... 63

Figure 5.1: Schematic of the fabrication steps required to obtain metal electrodes on glass

substrate. ........................................................................................................................... 65

Figure 5.2: (a) Schematic of the electrical and microfluidic components of the device and

(b) photograph of the PDMS microchannel with co-planar electrodes. ........................... 66

Figure 5.3: Transverse electrophoretic migration of pUC19 vector in 1× TBE buffer

(bottom-up view). (a) No field is applied, (b) when 1 VDC was applied, (c) 2 VDC and (d)

3 VDC. ................................................................................................................................ 71

Figure 5.4: Fluorescence intensity profiles of pUC19 (a), and -phage DNA samples (b)

in 1× TBE buffer across the channel width and at different times after the field is applied

showing the transverse DNA migration. ........................................................................... 72

Figure 5.5: pUC19 electrophoretic migration and concentration in 1× TE buffer (bottom-

up view). (a) No field is applied, (b) when 1 VDC was applied, (c) 2 VDC and (d) 3 VDC. 74

Figure 5.6: Fluorescence intensity profiles of pUC19 (a) and -phage DNA samples (b)

in 1× TE buffer across the channel width and at different times after the field is applied

showing DNA trapping and concentration at the channel center. ................................... 75

Figure 5.7: Rhodamine labeled BSA migration using 1× TBE buffer and 1× TE buffer.

(a) image of BSA when no field was applied, (b) BSA in 1× TBE buffer when 3 VDC is

xiv

applied, (c) BSA in 1× TE buffer when 3 VDC is applied, and (d) fluorescence profiles

across the channel width. .................................................................................................. 77

Figure 5.8: DNA sample under alternative square wave 3 Vpeak-to-peak (bottom-up view). 79

Figure 5.9: Images at the main channel entrance of DNA samples in 1× TE buffer at

different flow rates (bottom-up view). (a) No field, 0.5 μl/min flow rate, (b-d) 3 VDC was

applied when the flow rate was (b) 0.5 μl/min, (c) 0.3 μl/min, and (d) 0.1 μl/min. ......... 80

Figure 5.10: Images of pUC19 sample in TE buffer as time increases under a pulsed

electric field of 15 VDC at a 1 MHz frequency and 20% duty cycle (bottom-up view). ... 81

Figure 5.11: Effect of buffer selection on pUC19 migration under transverse DC electric

field. (a) No field applied, (b-d) 3 VDC was applied while pUC19 was diluted in three

different buffers: (b) 1× TBE, (c) 1× TE, (d) 1× PBS. ..................................................... 83

Figure 5.12: 2D geometry representing the cross-section area of the channel. ................ 84

Figure 5.13: Measurements of zeta potentials for selected buffers using PDMS/PDMS

and PDMS/glass channels [101]. ...................................................................................... 85

Figure 5.14: EO mobility as function of ionic strength for ACES buffer using fused silica,

PDMS/glass and PDMS/ PDMS channels [62]. ............................................................... 86

Figure 5.15: TBE buffer case: (a) potential surface plot and field streamlines, (b) field

surface plot and velocity streamlines, (c) velocity field surface plot and arrow plot. ...... 89

Figure 5.16: TE buffer case: (a) potential surface plot and field streamlines, (b) field

surface plot and velocity streamlines, (c) velocity field surface plot and arrow plot. ...... 90

Figure 5.17: PBS buffer case: (a) potential surface plot and field streamlines, (b) field

surface plot and velocity streamlines, (c) velocity field surface plot and arrow plot. ...... 91

xv

Figure 5.18: Electroosmotic velocity profiles in z direction (Uz) at the center of the

channel and across the channel height for TBE, TE and PBS buffers. ............................. 92

Figure 5.19: Net DNA velocity profile with EO and EP components (UEO+UEP) across

the channel height for TBE buffer. ................................................................................... 93

Figure 5.20: Estimated Net DNA velocity profile with EO and EP components

(UEO+UEP) across the channel height for TE buffer.......................................................... 94

Figure 5.21: Snapshots of a helical flow formation using fluorescent uncharged melamine

and negatively-charged polystyrene beads suspended in TE buffer under transverse field.

........................................................................................................................................... 95

1

Chapter 1

Introduction

1.1 Significance of DNA Purification

Since recombinant DNA technology development, DNA has been recognized as an

important component for biological and biomedical research because of its ability to code

for proteins and enzymes required for cell life, replication and adaptation to changing

environments as well as establishing the genetic basis of disease. DNA-based studies and

genetic engineering have greatly benefited from DNA‟s chemical stability, ease of

manipulation and analysis. DNA is purified from cells (bacterial or eukaryotic) or tissue

lysates, cut into individual fragments using highly specific restriction endonucleases,

amplified via polymerase chain reaction (PCR) and analyzed by gel electrophoresis and

southern blotting techniques.

Nucleic acid purification is widely utilized in a large number of applications throughout

the biomedical sciences. Some of these applications are important health-related areas in

where detection of specific DNA fingerprints is required including: the determination of

genetic markers or gene mutations implicated in diseases such as sickle cell anemia or the

identification of certain gene alleles which increase the risk of certain cancers,

2

genotyping of viral infections in humans, such as influenza, human immunodeficiency

virus (HIV), or hepatitis C virus (HCV) to direct treatment options, as well as the

detection of biowarfare agents.

In addition to the biomedical sciences applications, the discovery of DNA has resulted in

tremendous advances in agriculture; for example, where it is being use to genetically

modify the seeds to create crops with attributes that they do not have naturally. By

manipulating plant genes, agricultural yields have been improved achieving much higher

productivity and pest-resistant varieties.

DNA can be also used to investigate historic and prehistoric events. By reconstructing

gene changes and homology, modifications in DNA sequences can provide useful

information to understand how a specie evolves in time. In addition, since mitochondrial

DNA is passed on through the maternal lineage, it has been used to trace the migration of

humans over the planet.

In addition, the use of DNA for identification of people is largely utilized in forensics

which represent fundamental criminal evidence as well as in paternity and maternity

disputes through genetic finger printing of specific DNA polymorphisms found in the

human genome. A particular and important example of DNA analysis is the development

of DNA-based identity test which has recently provided valuable information for

identification of missing people during catastrophes or massive attacks, such as the

identification of September 11th

victims.

An area that also became important is the use of DNA analysis for defense; DNA

analysis for rapid and unequivocal detection of potential biological warfare agents is a

3

major objective for military authorities and has been used to trace the anthrax spores used

in the 2001 attacks.

Also, another growing area for DNA extraction is nucleic acid-based pathogen detection

in processed and unprocessed food products [1]. The presence of pathogenic bacteria can

cost consumers and the food industry many millions of dollars every year due to food

recalls and can cause many deaths each year around the world. Therefore, the rapid

bacterial detection by DNA-based analysis could help stop distribution of contaminated

foodstuffs.

As can be recognized, there are multiple areas of applications that can benefit from

advances in DNA preparation techniques. With recent developments in microfluidic

technologies, standard DNA extraction protocols can be improved by using smaller and

faster devices which can be integrated with other processes such as downstream PCR

amplification or capillary electrophoresis (CE) separation [2, 3] in a more integrated,

automated and compact fashion to allow autonomous sample preparation and analysis

modules. By using modern engineering technologies developed for the microchip

fabrication in the last decades, it is now possible to make the process of DNA isolation,

purification and analysis much faster, less expensive and more portable.

Focusing on the biomedical applications, the healthcare sector can greatly benefit by

rapid nucleic acid-based tests when fast diagnostics play a crucial role in saving people‟s

lives. One example is the rapid pathogenic agent detection to avoid epidemics. In isolated

and poor regions, the portability and inexpensive characteristics of lab-on-chip

technologies can make a significant improvement in rural health clinics. The use of low

volumes in comparison with standard methods makes this emerging technology suitable

4

for situations where the biological sample available is scarce such as in prenatal and

neonatal DNA diagnosis. Finally, this new technology brings the possibility of

integration and automation of multiple and complex analytical processes required in

biological diagnosis into a single platform and allowing more reliable results.

1.2 The Need for Sample Pre-concentration

As mentioned above, the growing developments of microfluidic technologies in recent

decades have guided a paradigm shift in the way chemical and biological samples can be

analyzed. Significant research efforts have been undertaken to develop miniaturized

platforms to purify, amplify, separate and analyze DNA in microchannels, with additional

efforts considering other important analytes like proteins, cells, organelles, and bacteria.

One important challenge as assay scale decreases is the sample pre-concentration which

often must be investigated to improve the sensitivity of a microchip assay. Usually a pre-

conditioning of the sample needs to be conducted prior the sample analysis to assure

purity and a minimal concentration required to provide an unequivocal detection. Besides

the enhancement in the detection sensitivity, sample pre-concentration also improves the

reliability of analysis by increasing the signal-to-noise ratio.

Along these lines, numerous microfluidic approaches have been investigated for

concentrating dilute analytes for example prior to micro-capillary electrophoretic (μCE)

separations in order to increase the signal at the output of the device. In some

applications, pre-conditioning of samples is critical to increase an analyte concentration

above the detection limit or to improve sample injection reproducibility which can be

5

negatively affected by diffusion, hydrodynamic and electrodynamic broadening, Joule

heating, and electromigration dispersion [4].

Different researchers have suggested several mechanisms for pre-concentration. Some of

methods for both DNA and proteins have included using electrokinetic phenomena such

as electrophoresis, electroosmosis and dielectrophoresis. For example, microfluidic

devices have been developed to investigate electrophoretic sample stacking techniques

including field-amplified sample stacking (FASS), isotachophoresis (ITP), and pH-

mediated stacking which are one successful example of the necessary preconditioning

step prior an efficient electrophoretic sample analysis. In addition to this, other

concentration mechanisms such as the use of specific chemical binding of DNA to a solid

phase or utilization of size exclusion membranes have been also exploited. A review of

these methods can be found in Chapter 4.

The trapping and concentration of analytes are needed in multiple biomedical

applications. The most extended use is as μCE preconditioning step that helps to achieve

minimal concentration limits and enhance signal-to noise ratio prior the electrophoretic

separation and analysis. Another important use is to maintain the sample of interest

concentrated and separated from contaminants or undesired substances. For instance,

electrokinetic methods have been used to isolate and sort cell components [5].

Additionally, having the sample locally constrained could be useful to enhance chemical

reactions as other chemicals are infused conjunctly.

6

1.3 Dissertation Overview

In this work two main ideas have been explored. The first part of this dissertation refers

to the design, fabrication, and performance of microfluidic devices used as platform for

DNA extraction by organic solvents. The main aims were to quantify protein partitioning

between the aqueous and the organic phase and DNA purification within the device, and

to evaluate two modes of extraction, passive diffusion through stratified flows and

droplet enhancement of extraction efficiency. The second part of this work is focused in

the behavior of DNA and other analytes during the application of transverse electric

fields. The main aims for this section were to understand the forces involved in the

electrokinetic migration of DNA under transverse fields and to investigate the buffer

solution influence in the analyte migration and concentration.

In chapter one, an introduction to the DNA extraction significance in the biomedical field

and the need for sample pre-concentration are addressed to understand the importance

and relevance of the research.

The second chapter presents an overview of the DNA extraction methods, a state-of-art

review of recent research towards DNA extraction chips, the theoretical framework for

two-phase systems in microfluidic devices, the partitioning mechanism and lastly,

different mixing methods to enhance protein partition.

Chapter three examines the criteria used to design the DNA extraction devices, the

fabrication process and the materials used in the experiments followed by the results

obtained from experiments. Protein partitioning and DNA purification were analyzed for

stratified flows and droplet-based flows in this section. A DNA recovery comparison

7

between the designed microfluidic device, standard phenol extraction and a commercial

extraction kit was made by using bacterial lysates and summarized in this chapter.

In chapter four, as an introduction to the second part of this dissertation, an introduction

to electrokinetic theory is provided which will be useful to understand the trapping

mechanisms of the sample pre-concentration module presented in Chapter 5. In addition,

a review of different sample trapping and concentration enhancement methods is

presented.

Chapter five contains the fabrication methods of the device proposed including the

embedded electrode traces manufacture. The sample and buffers utilized are detailed in

this chapter as well as the buffer characterization. The results obtained using transverse

electric fields and the different trapping mechanisms are explained in this section.

In chapter six the results obtained for DNA extraction via phenol extraction and sample

pre-concentration using transverse electrokinetic forces are summarized and discussed.

Finally, future prospects for improving this research and steps required towards creating

integrated DNA extraction and analysis devices are addressed.

8

Chapter 2

Background of DNA Liquid Extraction

2.1 Standard DNA extraction methods

DNA can be separated and purified by two leading methods; by using organic solvents,

also called phenol extraction, and by solid-phase-based extraction techniques. A

description of these two methods is provided:

Phenol Extraction

Phenol extraction is an organic-aqueous liquid extraction technique that separates DNA

and RNA from cellular components and proteins based on their affinity to different

immiscible solvent phases. The procedure consists of the digestion and lysis of cells or

tissue, mixing of the phases where proteins and lipids segregate into the organic phase

while DNA stays in the aqueous phase, microcentrifugation, separation and precipitation

of DNA with ethanol [6].

First, the cell lysis takes place which is a process of breaking open cells to release the

DNA into the solution and leave it accessible. Typical methods include the use of

9

protease enzymes, chemical agent such as detergents, or mechanical disruption of cell

membranes.

Second, the extraction is usually prepared in a 1.5 ml microcentrifuge tube with a

minimum volume of 100 μl required for easy manipulation with micropipette tools. The

aqueous sample with the lysed cell extract and an equal volume of

phenol:chloroform:isoamyl alcohol (25:24:1 v/v) are incorporated into the tube and

vortexed vigorously for 5 minutes to mix the phases. During this process, the cell

components distribute into either the aqueous phase or the organic phase in order to

minimize interaction energies. The cell membrane lipid components and proteins will

partition to the organic phase and DNA will stay in the aqueous phase due to its polar

nature having greater affinity to the aqueous phase. The separation occurs at the organic-

aqueous interface. Increasing the interfacial interaction area is central to maximize the

effective partitioning. In macroscopic systems this is accomplished through the vortex

mixing.

Following partitioning, centrifugation is the next step to separate the two phases by

density differences. The aqueous phase, the upper one in the tube due to its lower density,

is now removed with a micropipette tool. This extraction and centrifugation process can

be repeated to remove more components and to completely separate the phases.

The aqueous phase is then mixed with 2.5 volumes of ice cold 100% ethanol, and

microcentrifuged at high speed which precipitates the DNA into a pellet. The supernatant

solution is removed. Finally, 70%-80% ethanol is added, mixed and discarded to wash

the remaining pellet removing excess salts. After washing, the DNA is resuspended in

50-100 μl of Tris-EDTA (TE) buffer.

10

Solid Phase Extraction (SPE)

This technique is based on the absorption of DNA onto a silica solid phase under high

chaotropic salt concentration condition to separate DNA from the cell components. Many

purification kits are commercially available (e.g. Qiagen, Promega, Biorad, Epicentre).

Commercial kits usually utilize solid phases like silica gel, glass matrix, or membranes to

bind and purify DNA. The main steps for solid phase extraction are lysis of the cells,

absorption of DNA, washing, and elution of DNA [7].

Cell lysis is done under alkaline conditions and the solution is then loaded into the

purification kit which consists of a packed silica column. In presence of high

concentration chaotropic salts, the DNA binds to the silica surface because of

electrostatic shielding between the DNA and the silica surface while other cell

components do not bind.

A washing step is required after the DNA adsorption to flush proteins and other

contaminants from the column. Via the introduction of an elution buffer, the purified

DNA is then recovered at low salt concentration which promotes the detachment of the

DNA molecules from silica.

When comparing these two techniques advantages and disadvantages to each procedure

can be found. SPE does not have the inconvenience of dealing with organic solvents like

phenol extraction. Residual phenol can denature protein and quench further chemical

reactions if phenol contamination is present. However, solid phase extraction is limited

by the exposed surface area to which DNA can attach to and the extraction efficiency

11

obtained with this technique is lower than the phenol extraction-ethanol precipitation

procedure and is more susceptible to protein contamination due to non-specific binding.

Furthermore, SPE is a fundamentally batch process requiring sequential introduction of

solutions over a fixed column, whereas liquid extractions have the possibility to be run

under continuous flow conditions.

2.2 Microfluidic DNA Extraction

The transfer of traditional analytical techniques into microscale platforms has allowed the

automation and optimization of analytical processes and point-of-care medical systems.

For this reason, integrating cell lysis, purification, amplification and analysis of nucleic

acids into a micrototal analytical system (μTAS) represents the driving goal of many

research groups all over the world. Microfluidic approaches towards integrating DNA

handling processes have focused mainly on miniaturization of solid phase extraction,

studying different materials, and microfabrication methodologies to bind DNA

efficiently. Some approaches are reviewed here:

Silicon pillars

In 1999, Christel et al. developed of a cartridge where lysed cells are introduced into a

DNA purification module followed by a PCR amplification module [8, 9]. The DNA

extraction chip consists of silicon pillars, made by deep reactive ion etching (DRIE),

where DNA binds (Fig. 2.1a). The use of these columns increases the surface area to

volume ratio, providing a high binding capacity which is approximately 40 ng/cm2.

12

Washing and elution step are incorporated to the chip as well as PCR thermal cycles to

amplify DNA quantities in cases of low concentration samples. The integrated cartridge

with the different process chambers is shown in Fig. 2.1b. The extraction efficiency was

found about 50% using bacteriophage λ DNA as the target and DNA extraction was

successful with both low and high concentrations of DNA samples (100-1000 ng/ml).

Figure 2.1: (a) SEM photograph of a pillar-based DNA extraction chip. (b)

Integrated microfluidic cartridge containing the pillar-based DNA chip and a PCR

reaction tube [8, 9].

Silica bead/sol-gel systems

Another solid phase that has been explored is silica beads in order to enhance the surface

area. In 2003, Landers‟s group proposed silica beads immobilized within a sol-gel instead

of silicon or glass pillars [10]. Beads packed in the microfluidic device increase the

absorption area where DNA will bind. The channel is first loaded with the beads and then

(a) (b)

13

filled with sol-gel to immobilize the silica particles (Fig. 2.2C) that involves the transition

of the solution form liquid (colloidal “sol”) into solid form (“gel”). In addition, to keep

the beads trapped in the device, the gel helps keep them separated, avoiding clogging,

and increases the effective absorption surface area. The chip consists of a simple channel

as shown in Fig. 2.2A. It requires an outside lysis step and external loading of sample,

washing buffer, and elution buffer. This device was tested with blood and bacterial

samples and has demonstrated an extraction efficiency of 50% for human genomic DNA

and 70% for λ DNA [10].

Figure 2.2: Silica particle-based microchip with sol-gel immobilization. A) 1x

magnification; B) 10x magnification; C) 500x magnification of cross section [10].

14

In 2004, Quake‟s group presented a fully integrated microfluidic device using silica

beads loaded previously to the chip where DNA samples are attached and eluted by the

use of pneumatic valves [11]. The DNA extraction is followed by PCR amplification to

achieve minimal concentration for analysis. The fabricated integrated chip was made

using soft lithography techniques and all the processes involved in the purification of

nucleic acids were integrated in one chip (Fig 2.3) [11].

Figure 2.3: (a)-(e) Schematic steps of bacteria lysis and DNA purification process. (f)

Image of the DNA microfluidic chip with parallel architecture [11].

By external pneumatic activation, valves can control the trafficing of the different buffers

and samples to the corresponding chamber, and isolate units from the rest of the chip.

The main units are lysis buffer chamber, dilution chamber, and cell chamber. In Fig 2.3a,

cell sample, dilution buffer and lysis buffer are introduced first into the device and

15

pumped through a rotary mixer which is also controlled by hydraulic valves that

constantly open and close the entrance of fluids. The function of this step is to mix the

components and reduce the time taken for lysis of the cells. Then, the mixture is flushed

over beads which are preloaded in the device. DNA binds the beads and a washing buffer

is used to flush the waste. Finally, the elution buffer is added and DNA is collected. In

this fully automatic fashion, DNA extraction can be performed with a very little user

handling.

Other novel phases

More recently, other phases have been investigated to address two main issues: increase

the surface area (loading capacity) of the phase and reduce the non-specific binding of

proteins. The use of beads increased the surface area but also increased the back pressure

within the device, while the use of DRIE silicon pillars involves a high microfabrication

cost. Some new alternatives to these original methods proposed by other researchers are

organic monolithic phases [12-15] which improved the extraction efficiency up to 70%.

The use of porous silicon devices [16] resulted also in high DNA recovery (~80) as well

as chitosan coatings [17], which reported to obtain extraction efficiencies within a range

of 70% to 85%. Finally, instead a silicon substrate, Witek and co-workers explored

genomic DNA purification from bacteria using photoactivated polycarbonate devices

with high recovery (~85%) and DNA binding capacity (~150 ng) [18].

16

Microfluidic Phenol Extraction

Zahn‟s group conducted the preliminary phenol extraction studies using microfluidics by

examining firstly the organic-aqueous interface stabilization [19]. Furthermore,

experimental and computational evaluation was made by using droplet-based flows

(Figure 2.4) as a proposed mixing procedure for DNA extraction [20].

Electrohydrodynamic instability (EHD) mixing [21] and diffusion limited approaches to

protein and DNA partitioning [22] were also addressed. Lately, droplets-based flows

were utilized to extract DNA plasmid from lysate solution and compared with standard

extraction methods, achieving ~92% DNA recovery. These recent developments of the

microfluidic phenol extraction, compiled in a journal paper [23], are part of this work and

are discussed in detail in Chapter 3.

Figure 2.4: Photograph of aqueous droplet formation containing cell lysates,

including green fluorescent protein GFP which partition to organic phase [20].

17

2.3 Theoretical background: Mechanism of Partitioning

In a two-phase system, partitioning consists of a selective distribution of solute according

to their properties [24]. The solutes are affected by multiple forces including hydrogen

bonding, ionic, hydrophobic, van der Waals and other weak forces with the surrounding

fluid phase and interactions with other molecules which affect the transport [24]. The

contribution of each force is difficult to calculate; however, the net effect can be

characterized by the partition coefficient:

1

2

CK

C (2.1)

where C1 and C2 are the concentrations of the partitioned molecule at equilibrium in each

phase [24]. The partition coefficient is function of the properties of the two phases,

molecular characteristics, such as size and charge, and temperature, but ideally

independent of concentration [24].

In a two-phase system, the distribution of the molecules is driven by two main

phenomena, the thermal motion of the molecules that distributes them in each phase, and

the partitioning forces, mentioned previously, which move the solute from one phase to

the other. These interactions can be related as:

/1

2

E kTCe

C (2.2)

where E is the energy required to move a molecule from phase 1 to phase 2, k is the

Boltzmann constant, and T is the temperature [24].

18

2.4 Theoretical background: Two-Phase Microfluidic Systems

Multiphase flow is defined as two or more different immiscible fluids present in a

system. Multiphase flows at microscale are complex because the flow involves multiple

forces that are not significant at macroscale dimensions. However, these systems have

several attractive characteristics which make them very practical for applications in

microfluidics [25]. In recent years, many efforts have been done to understand multiphase

fluid behavior and to find methods to control droplet-based and stratified flows, as well

as several applications of these systems [25-30].

The description of fluid dynamics at small scales is quite different if it is compared to

large-volume systems [31]. The Navier-Stokes equations and the continuity equation

describe the fluid physics in the two bulk fluids, assuming conditions of constant density

ρ and viscosity μ and incorporating the interfacial effects γ:

2pt

uu u u f (2.3)

0 u (2.4)

where u is the velocity of the fluid, p is the pressure and f represents body force density.

In microscale systems, the inertial forces are much smaller compared to viscous forces

and the nonlinear term in (2.1) can be then neglected, leaving the Stokes equation [25,

31]:

2pt

uu f (2.5)

19

In this work presented here, the driving force is always pressure provided by a syringe

pump but other external forces can be applied to microfluidics such as electrical,

magnetic and thermal [31, 32].

In addition to the bulk fluid forces, the force equilibrium condition at the interface

located between the two phases requires that

ˆ( ) 0A C

dA dl T T n t (2.6)

where ˆand T T are total bulk stress in the flow above and under the surface element A,

respectively, n is the unit vector normal to the interface, γ is the interfacial tension, and t

is the unit vector tangential to the interface and normal to the boundary curve C.

Applying Stokes‟s theorem to (2.6) and considering an arbitrary surface element A, it

follows that

ˆ( ) ( ) 0grad T T n n n (2.7)

To obtain the normal stress balance, inner product is applied to (2.7) and T is replaced by

p ij ij ijT δ where p is pressure and τ is the shear stress component

j i

i jx x

ij

u u , this gives

ˆ ˆ( ) [(( ) ) ] ( ) 0tot totp p n n n (2.8)

Equation 2.8 states the relationship between pressure, shear stress, interfacial tension and

curvature at the intersection of the two phases. The changes of interface curvature

compensate for the differences in pressure and shear stress in both bulk phases to

maintain the stress equilibrium at the interface.

20

In fluid mechanics it is common the use of dimensionless numbers which relate the

competition between different forces. In Squires and Quake‟s review [31], a complete list

of dimensionless number used in microfluidics is shown. However, if we consider the

following forces: buoyancy, gravitational, inertial, viscous and interfacial forces, the

importance of each force will dramatically change as the scale goes from the macro to

microscale. In microfluidics, buoyancy, gravitational and inertial forces become less

important and can be neglected. On the other hand, viscous and interfacial forces, usually

ignored in the macroscale, become very important to explain the fluid behavior in

microchannels [25].

The Reynolds number is the most representative number to characterize flows and it is

the ratio between inertial and viscous forces:

ReUL

(2.10)

where U is the average flow velocity and L is the characteristic flow dimension [32]. In

microscales, Re for liquid flows usually vary between the order of 10 and 10-4

[25].

Therefore, the inertial forces can be ignored and only laminar flow is expected. In

addition, in simple channels and pressure-driven flows the velocity profiles are parabolic.

More about fluids under low-Re-number conditions can be found in the literature [33,

34].

In multiphase systems, the capillary number is the most significant dimensionless

number. It relates the viscous forces which act tangential to the interface elongating it and

the interfacial forces which act normal to the interface minimizing the interfacial area by

inducing droplet formation [19]. The capillary number can be expressed as:

21

UCa

(2.11)

If the viscous forces are more significant than the interfacial forces, the multiphase flow

behavior is parallel or stratified (Ca>>1). Segmented or droplet-based flow occurs when

interfacial forces are predominant (Ca<<1). Other regimes appear in the transition such as

pearl-necklace and pears regimes [35, 36]. The analysis of multiphase flow behavior has

been done in multiple channel configurations such as T, Y and cross junction

configurations [25].

Multiphase flow pattern are influenced by multiple factors. The fluid behavior depends

on interfacial forces between the fluids and channel walls, individual fluid viscosities,

fluid velocities, and the geometric characteristics of the channel [35, 37-39].

First, interfacial forces can be altered using surfactants which will stabilize the interface

due to the reduction of the interface tension [19]. Wall surface treatment, guide-structure

microchannels and self assembled monolayer modifications are other methods to control

flow behavior in microchannels [40-42].

Second, viscosity can be chosen to influence flow patterns in a desirable way. Stratified

flows are more frequent with high viscosity fluids or high viscosity ratios between fluids

[35, 39]. The fluid viscosity can be altered by adding components to make it more

viscous such as polymers, or by changing temperature [25].

Third, the most widely used method to control multiphase flows is by tuning fluid

velocity. Pumping methods used in microfluidics can easily modulate flow rate that

control fluid behavior and droplets sizes [27, 28]. By increasing the flow rates, the flow

pattern will become stratified due to prevalence of viscous forces [20, 35, 36, 38]. In

22

addition, when velocity difference increases, the shear forces to normal forces ratio

increases and segmented flow turns into stratified flow [35, 37].

And finally, the flow behavior can be also influenced by channel geometry. It has been

shown that channels with higher aspect ratios enhance droplet formation [37]. There is no

report of the influence of the channel length. The entrance configuration is also another

parameter to take into consideration to enhance one behavior or the other and to control

droplet size [25, 27, 28, 43].

All these factors give the framework that designers will use to delineate multiphase

microfluidic systems. The aims of many researchers are to achieve reproducibility and

fully control of microscale multiphase flows to use these systems in multiple

applications. Some examples of multiphase applications are: microreactions and

molecular transport (enzyme kinetics, protein crystallization, syntesis of nanoparticles)

[30, 41, 42, 44, 45], solvent extraction [19-21, 40], drug encapsulation and mixing [25,

27, 28].

2.5 Convective Transport Enhancements

In microfluidics, one of the major challenges is to achieve rapid and effective mixing.

This is because at microscales the flow is laminar and mixing is usually diffusion limited.

The Péclet number (Pe) is defined as

LUPe

D

(2.12)

23

where L is the channel characteristic diffusion length (here assumed to be the half width

of the microchannel corresponding to the diffusion length of protein from the aqueous to

the phenol phase), U is the average fluid velocity and D is the protein diffusivity. When

considering mixing in microfluidic channels using stratified adjacent flow streams, the

Péclet number is usually very high implying long channel lengths and long diffusion

times to achieve full mixing.

In the case of two-phase systems, both diffusion within the phase and diffusion across the

solution interface need to be maximized. Next, two different techniques that could be

easily adapted to enhance mixing in a multiphase system are discussed.

Mixing enhancement using droplets

Rapid mixing with droplets can be achieved due to the convective enhancement within

droplets produced by two recirculating flows inside of each phase. Figure 2.5 shows the

two flows generated in a droplet as it moves forward. As the droplet moves, the fluid

interface needs to recirculate or „tank-tread‟ and this shear is transported into the droplet

causing the recirculation vortices. Two recirculating flows are also present in the

continuous phase but they are not shown in the figure. These circulating flows mix the

molecules within each phase and also increase the diffusion through the interface by

enhancing the molecule mass transport to the fluid-fluid interface [20, 30]. In addition to

convection enhancement, droplets increase molecule partitioning across interfaces due to

the total interface surface area is being increased.

24

Figure 2.5: Illustration of multiple reagent mixing in a droplet by recirculating

flows [30].

Chaotic Advection

Chaotic advection mixers are three-dimensional structures that mix the flow by

reorienting the flow alternatively which produce the stretching and folding of the fluid

stream. Three-dimensional serpentine microchannels (two-layers device) have been

developed with significantly mixing improvements [46]. In addition, Stroock et al

designed a chaotic mixer by using asymmetric herringbone-like grooves that produce a

three-dimensional twisting flow [47].

In the case of a two-phase system, the plugs already have inside a two-dimensional flow

due to the internal recirculation. Mixing can be easily enhanced by using serpentine-liked

structures (three dimensional twist) without the requirement of two-layer fabrication. In

each direction change, the mixing inside of each phase is enhanced due to the stretching

and folding of the droplets [30, 48, 49].

Both stratified and droplet-based flows benefit by this passive mixing. In Figure 2.6,

droplet enhancement and chaotic advection mixing strategies are combined showing how

25

recirculating flows are reoriented from the direction of the droplet movement, streached

and fold, resulting in a more efficient mixing.

Figure 2.6: Droplets moving through a winding microchannel, (a) experimental

results and (b) schematic of the droplet folding [50].

This review of the theoretical aspects of multiphase microfluidic system provides the

central guidelines for the design of the DNA purification module addressed in the

following chapter.

26

Chapter 3

Microfluidic DNA purification via phenol extraction

In this chapter, microfluidic platforms are described and tested using organic solvents to

investigate the protein partitioning while DNA can be extracted from an aqueous phase.

Device design, fabrication and materials used for the extraction are presented followed by

the results obtained for bacterial lysates using microfluidic platforms, standard macro-

scale method for phenol extraction and a commercial DNA purification kit.

3.1 Device Design

During phenol extraction, the cell components distribute into either the aqueous phase or

the organic phase to reduce interaction energies with component partitioning occurring at

the organic-aqueous interface. Therefore, maximizing the interfacial interaction area is

critical to effectively partition the different cellular components. In macroscopic systems

this is accomplished through mixing using a vortex mixer to create chaotic and vortical

flow fields interspersing the two fluid phases. Within microfluidic systems, these types of

flow profiles cannot be easily generated so mixing and partitioning must rely on diffusive

mass transport or droplet interspersion of the phases as described in previous chapter.

27

When the two phases are infused as a stratified flow the flow rate must be large enough

so that a stable stratified flow can be maintained, while the residence time of the fluid

within the microchannel must be long enough for complete diffusive mixing. As has been

long recognized in the microfluidics field, when the two fluid streams flow at a high

linear velocity (~mm/s or greater) minimal diffusive mixing between the streams is

expected. In order to design the microchannel three parameters were considered: the

Péclet number, capillary number and diffusion time required.

The Péclet number LU

PeD

relates the convective transport of the material flowing

down the length of the channel (LU) to the transverse diffusion of proteins (D) across the

channel width from the aqueous phase to the organic-aqueous interface. In microfluidic

platforms the Pe is usually very high (~100 or greater); while the length of channel

required to ensure full diffusive mixing varies linearly with Pe, implying a long

microchannel is required.

The operating flow velocity, U, used in this study is constrained by the interfacial forces.

In microfluidic multiphase systems, the flow patterns are determined by a variety of

factors such as fluid choice, flow velocity, viscosity ratio, interfacial tension, device

wettability and geometry of the channels, as addressed in Chapter 2.

One of the most important factors in determining multiphase flow structure has relied on

the capillary number c cUCa

, which relates the relative importance of viscous shear

forces on the continuous fluid phase (μcUc) which act tangentially to the interface,

elongating it with the interfacial forces (γ), which act normal to the interface, minimizing

the interfacial area by inducing droplet formation [19]. As the viscous forces become

28

more significant than the interfacial forces, the multiphase flow behavior is parallel or

stratified (Ca~1 or greater). Segmented or droplet-based flow occurs when the interfacial

forces are dominant (Ca<<1).

The preliminary microfluidic device design was made considering a stratified flow

condition assuming Ca is equal to 1 which constrains the fluid velocity. The diffusion

time required for the protein to reach the organic-aqueous interface was estimated as:

2

4D

wt

D

(3.1)

where w is the half width of the microchannel and D is the diffusion coefficient of the

molecules to be extracted. The diffusion time was calculated assuming w equal to 30, 40

and 50 m and a BSA diffusivity in water equal to 5.9x10-11

m2/s [51]. By setting the

required diffusion time equal to the fluid residence time inside of the device, the

microchannel length was calculated using the fluid velocity fixed by the capillary number

condition where the required channel length, channelL is equal to

442 wPeDwUtU D . The channel lengths were 41.8, 57 and 64.8 cm, for the

corresponding half width of 30, 40 and 50 m, respectively.

Based on these assumptions for passive diffusion across a stratified flow, long serpentine

devices were first fabricated to investigate protein partitioning into an organic phase

using both a stratified flow and droplet-based flow. A second device was also fabricated

after the results using the long serpentine and the design modifications were addressed in

the next section.

29

3.2 Device fabrication

All microchannels designed were fabricated using the soft lithography approach as

described in the literature [52]. The microchannel designs were patterned onto a silicon

master using standard photolithography techniques followed by casting of

polydimethylsiloxane (PDMS) and bonding to glass. A schematic of the photolithography

and soft lithography techniques is shown in Figure 3.1 where each step of the fabrication

is presented. A detailed protocol for the device fabrication is provided in Appendix A.

Figure 3.1: Fabrication steps of photolithography and soft lithography techniques.

The master mold was lithographically patterned on a spin coated SU-8 negative

photoresist (MicroChem, Newton, MA). The PDMS structure was cast by mixing the

elastomer base and curing agent (Sylgard 184 Silicone Elastomer Kit, Dow Corning,

Midland, MI) at a ratio of 10:1 and poured on top of the master mold. After degassing in

30

a vacuum chamber, the PDMS cast was cured in an oven at 65˚C for 1 hour. The PDMS

structure was then irreversibly bonded to a glass slide following surface activation of the

PDMS and glass using a corona discharge generator [53].

The device was connected with 0.254 mm ID Tygon tubes (Small Parts Inc., Miami

Lakes, FL) and sealed with glue (J-B KWIK, Sulphur Springs, TX), specially selected

due to their compatibility with the organic phase. The serpentine devices consisted of a

long microchannel where the two phases were co-infused into a converging channel

geometry with channel width and length described previous section and height of 20 m

(Figure 3.2).

Figure 3.2: (a) Serpentine mask designs and (b) photograph of the microfluidic chip.

(b)

(a)

31

In experiments using droplet-based flows, the channel height was increased to 40 m to

promote droplet formation. A second design was developed to be used with bacterial

lysates and under only droplet-based flows. Due to the mass transfer enhancement

enabled by droplet flows, the total length of the device could be reduced to 10 cm, while

the channel width and height were 100 and 40 m, respectively, and the entrance

configuration of the microchannel was changed from a Y-type to T-type inlet to facilitate

the droplet formation.

3.3 Materials and Experiment setup

Aqueous phase preparation

The aqueous phase consisted of phosphate buffered saline (PBS) (150 mM NaCl, 8.4 mM

Na2HPO4·7H2O 1.8 mM NaH2PO4·H2O pH 7.5) (HyClone, Thermo Scientific, Logan,

UT) with DNA and protein analytes introduced to investigate protein partitioning and

DNA extraction. The analytes were bovine serum albumin protein (BSA) (Omnipur,

fraction V; EMD Biosciences) conjugated with rhodamine fluorescent dye and unlabeled

2-log DNA ladder (0.1 to 10.0 kb; New England Biolabs, Beverly, MA).

The conjugation of BSA to 5-6-carboxy-tetramethylrhodamine (TAMRA) succinimidyl

ester dye (Fluka, Switzerland) was made at a 10:1 molar ratio of dye to protein, and the

complete protocol can be found in Appendix B. The labeled protein solution was then

filtered through a PD-10 desalting column (Sephadex G-25M; GE Healthcare) to remove

the unconjugated dye from the solution. The output fraction with the labeled proteins was

collected, stored at -20°C and diluted in PBS prior to use.

32

Bacterial lysates preparation

E. coli HB101 K-12 (Bio-Rad, Hercules, CA) was transformed using a Bio-Rad

Transformation kit to insert a green fluorescent protein coding pGLO (5.371 kb) plasmid.

The bacteria colony was cultured at 37 °C for 14 hours in Luria-Bertani liquid medium

with 100 μg/μl ampicillin and 2% arabinose (Teknova Inc., Hollister, CA). Following

culturing, the bacterial cells were pelleted in 1.5 ml Eppendorf tubes by centrifugation at

10000 rpm at 4°C for 15 min. The bacterial lysates were prepared using Qiagen plasmid

purification buffers (Qiagen, Valencia, CA) as indicated in the Qiagen Plasmid

Purification Handbook [7]. See Appendix B for bacteria transformation, plate and liquid

culture protocols.

Bacterial pellets were resuspended in 300 l Qiagen buffer P1 (50 mM Tris-Cl, pH 8.0;

10 mM EDTA; 100 g/ml RNase A) and an equal volume of lysis buffer P2 (200 mM

NaOH, 1% sodium dodecyl sulfate (SDS) (w/v)) was added and incubated for 5 minutes.

Finally, 100 l of neutralization buffer P3 (3.0 M potassium acetate, pH 5.5) at 4°C was

added and incubated for 5 minutes on ice to precipitate large debris and SDS. After

centrifugation at 10000 rpm for 10 minutes, the supernatant was removed and used in

three tested experimental conditions; the microfluidic DNA extraction as well as in a

standard phenol extraction and Qiagen plasmid purification kit for extraction comparison

purposes.

33

Experimental procedure

The infusion of the aqueous and organic phases, contained in glass syringes (Hamilton,

Reno, NV), was made by using syringe pumps (PicoPlus 22, Harvard apparatus,

Holliston, MA) to control the infusion flow rates. The aqueous phase consisted of a PBS

buffer solution with the addition of 0.5% (w/v) SDS (J. T. Baker Chemical Co.,

Phillipsburg, NJ), in order to reduce the interfacial tension between the aqueous and

phenol phases and achieve a stable stratified interface between the phases [19].

The BSA protein concentration was diluted to 0.3 g/l in PBS, which had sufficient

fluorescence intensity for flow visualization while preventing precipitation of protein

within the microchannel. Protein precipitation was seen using BSA concentrations of 0.5

g/l or higher, causing obstruction of the flow within the microchannels. The final DNA

ladder concentration was diluted to 0.075 g/l.

The phenol phase was a mixture of three organic components: phenol, chloroform, and

isoamyl alcohol, at 50, 48 and 2% by volume concentration, respectively purchased from

Pierce (Rockford, IL) and used as received. The organic phase viscosity was previously

determined to be 3.52 x 10-3

Pa-s and the interfacial tension with the aqueous phase was

0.1 mN/m [19].

To analyze the fluorescence intensity of rhodamine in the channel, a Nikon Eclipse

TE2000U inverted microscope (Nikon, Tokyo, Japan) operating in epifluorescence mode,

a TRITC filter cube and charged coupled device (CCD) camera (PowerView 1.4 MP, TSI

Incorporated, Shoreview, MN) were used to acquire images. All the images were

acquired using a 10× objective, neutral density filter ND1 for the epifluorescence arc

34

lamp and a 26 ms exposure time. The average intensity of the fluorescently labeled BSA

in the aqueous phase at the entrance and at the outlet of the device were calculated using

ImageJ software, following subtraction of the average image intensity from each image

background, and compared to evaluate the protein depletion.

Additionally, spectrophotometric analysis was used for protein partitioning quantification

(Appendix B). Aliquots from the aqueous phase at the inlet and from the aqueous phase

collected at the outlet were measured using a 1 ml cuvette and a DU 730

spectrophotometer (Beckman Coulter, Fullerton, CA), at a dilution factor of 6. The

protein depletion was evaluated comparing rhodamine absorption at 555 nm of the

aqueous inlet and the aqueous outlet aliquots.

Finally, DNA extraction was quantified by gel electrophoresis analysis using 1% agarose

gel in 1× Tris-acetate-EDTA (TAE) buffer (see Appendix B for gel electrophoresis

protocol). Images of gel were taken by BioDoc-it Imaging System (UVP, Upland, CA)

and ImageJ was used to quantify extraction efficiencies comparing total intensity of the

DNA fragments of the inlet and outlet solutions. For all data presented here a set of four

experiments were used for statistical analysis.

3.4 Results and Discussion

Review of serpentine device comparison

As part of previous work (Master of Science Thesis), the serpentine designs were

compared based on their BSA depletion efficiency. The first experiments were realized

using the largest device (device III in Figure 3.2a, 100 μm wide and 64.8 cm long). For

35

this particular geometry, the parallel flow pattern was achieved in the entire serpentine

structure only at high flow rates (3 μl/min for the aqueous phase and 1.3 μl/min for the

phenol phase).

Figure 3.3: Comparison of the serpentine device design based on the protein

depletion. (a) using epi-fluorescent image analysis and (b) absorption spectra

analysis.

(a)

(b)

36

The resulting mean velocity of the aqueous phase was 5 cm/s. When the velocity was

lowered, the spontaneous generation of droplets at the entrance or in other sections of the

device was observed.

The comparison of protein extraction efficiency between the three different devices was

conducted at the same mean velocity (5 cm/s) in the aqueous phases for each of them.

The aqueous flow rates were 2.4 μl/min and 1.8 μl/min for device II and I, respectively,

with flow rates for the phenol phase of 1 μl/min and 1.3 μl/min in the same devices to

achieve a stratified flow profile.

Images of the phases were taken at different positions along each device and analyzed,

obtaining the average intensity of the aqueous phase at those positions and normalized to

the average intensity at the inlet. The BSA depletion comparison in each device is shown

in Figure 3.3a as a function of downstream position. As expected, the protein depletion in

the aqueous phase is faster in the narrow channel device (device I) than in wider ones

(device II and III). The partitioning efficiencies using the fluorescence intensity method

were 78%, 74.3% and 68.5% for device I, II and III, respectively.

The aqueous phase outputs from the three devices were collected and quantified by

spectrophotometry. From the spectra of the inlet and the outlet solutions in Figure 3.3b,

the high absorption at low wavelength was found in all the aqueous outputs after leaving

the device which is due to phenol contamination forming small micelles at the interface.

Using the absorption ratio at 555 nm between the inlet sample and the outputs solutions

from the three devices, the partitioning efficiencies were 90%, 78% and 69% for device I,

II and III, respectively.

37

Since the diffusion time is proportional to the square of the diffusion path from one phase

to the other (half width of the channel), as the microchannel width decreases the required

diffusion time decreases as well. From the device comparison presented above, device I

appears to be the most efficient at partitioning protein due to the small channel width;

however, reduced channel width can lead to other problems like high operating pressures

or channel obstruction by the accumulation of protein precipitates or cell debris. For the

following experiments, device II was chosen for flow comparison tests to avoid clogging

due to narrow inlets as in device I.

Review of flow pattern comparison

Once the device selection was made, two flow conditions, stratified or parallel flow and

droplet-based flow, were tested to investigate the protein partitioning into the organic

phase.

To produce a stable stratified flow profile with each fluid phase being half the channel

width, a flow rate of 2 l/min for the aqueous phase and 0.6 l/min for the phenol phase

were utilized resulting in a capillary number Ca of 0.72.

38

Figure 3.4: Serpentine device: (a) images of BSA partition into organic phase using

stratified flow, (b) the device geometry and dimensions, and (c) images of BSA

partition using droplet-based flow.

Using the CCD camera, images at different positions along the length of the device,

shown in Figure 3.4a, were taken to quantify the fluorescently labeled BSA in each phase

and evaluate the partitioning from one fluid phase into the other. According to

epifluorescence microscopy analysis and using a set of four experiments, 78 ±11.57% of

the rhodamine labeled BSA protein was removed from the aqueous phase through the

device length by passive diffusion.

During these experiments, the multiphase outputs were collected and analyzed by

spectrophotometry to evaluate the total partitioning, resulting in 78.4 ±15.2% depletion

Flow direction

(b)

(c)

Height 40 m, width 80 m, length 57 cm

Aqueous droplets

Aqueous phase

Organic phase

Rhod-BSA

80 m

Flow direction

Organic phase

Aqueous phase

Rhod-BSA

(a)

39

according this method. Absorption spectra of the inlet and outlet solution are shown in

Figure 3.5a.

Figure 3.5: Analysis of inlet and outlet aliquots for stratified flow using (a) dye

absorption spectra and (b) DNA gel electrophoresis (1: inlet, 2: aqueous outlet, 3:

organic outlet).

On the other hand, the high DNA recovery in the aqueous outlet was found using gel

electrophoretic analysis and the recovery using this type of flow resulted in 89.3 ±7.5%

(Figure 3.5b).

However, working with multiphase systems in microchannels allows the formation of

droplets which could have several advantages for DNA purification. First, the use of

droplets increases the interfacial area between the phases so proteins can quickly partition

into the phenol phase. Second, rapid mixing within the droplets can be achieved due to

the convective enhancement produced by two recirculating flows seen inside of each

(a) 1 2 3 (b)

40

droplet, which mixes the analytes within the aqueous phase which in turn improves the

transport to the organic-aqueous interface [20, 54, 55].

Figure 3.6: Analysis of inlet and outlet aliquots for droplet-based flow using (a) dye

absorption spectra and (b) DNA gel electrophoresis (1: inlet, 2: aqueous outlet, 3:

organic outlet).

Droplet formation was investigated using the same serpentine platform and BSA-DNA

mixture used to study partitioning and extraction under stratified flow. In this case, the

flow rates used were 0.4 l/min for the aqueous phase and 0.5 l/min for the phenol

phase at a Ca of 0.07. Fluorescence images at the entrance and at the outlet of the device

are shown in Figure 3.4c. It can be seen from these images that the protein partitioning

was greatly improved by the convective transport enhancement from the droplet based

flow compared to the experiments using stratified flow conditions. The protein

partitioning by fluorescence microscopy analysis was 91.6 ±5.1% and confirmed via

spectrophotometry analysis with a depletion efficiency of 96.27 ±1.73% (Figure 3.6a).

(a)

1 2 3

(a) (b)

41

The DNA recovery using droplet-based flow was calculated by gel electrophoresis

analysis (Figure 3.6b) as in the previous stratified flow experiments and the percentage

recovered was 97.4 ±1.32%.

Plasmid DNA Purification from Bacterial Lysates

Based on the information collected during the droplet-based experiments, the serpentine

device, originally designed to perform extraction using stratified flows, was replaced by a

shorter device (Figure 3.7) which was redesigned for segmented (droplet) flow. In the

previous experiment, droplets were shown to enhance mixing, removing the BSA from

the aqueous droplets to produce clear plugs, at a downstream distance of approximately

10 cm from the entrance of the device.

This finding allowed a dramatic reduction in the length of the device; and consequently a

residence time reduction. In addition to the channel length shortening, the device entrance

was modified to a T-type junction to promote the droplet formation. T-type channels have

been used by other researchers to investigate droplet formation [35-39], and it allows for

the generation of stable droplets more easily than the Y-type entrance due to the

crosschannel blocking and pinch-off mechanisms of droplet formation observed by other

research groups [36, 37].

42

Figure 3.7: (a) Redesigned device geometry and diamensions, (b) photograph of the

new shorter device and (c) images of rhodamine-labeled BSA partitioning from

aqueous phase into organic phase.

Having demonstrated the device‟s capability to extract DNA from a protein solution, the

next step was to perform plasmid DNA extraction from a bacterial culture and compare

the purification efficacy with other traditional methods. The shorter redesigned device

was utilized in this case, using bacteria lysates from transformed E. coli containing a

green fluorescent protein coding pGLO (5.371 kb) plasmid.

The lysates solutions were prepared from a bacterial culture using Qiagen buffers for

lysis, as described in previous section. Next, the solutions were used in the microfluidic

device, Qiagen extraction column and standard macro-scale phenol extraction to compare

their extraction efficiencies.

(b)

Height 40 m, length 10 cm, width 100 m

(a)

Aqueous droplets

(c)

Lysate solution

Organic phase

Rhodamine dye

Rhodamine dye

43

The lysates were then infused conjunctly with the phenol phase into the microfluidic

device demonstrating successful recovery of plasmid in the aqueous phase, based on the

gel electrophoresis results presented in Figure 3.8a.

Figure 3.8: Comparison using the redesigned microfluidic device, Qiagen kit and

standard macroscale phenol extraction: (a) gel electrophoresis analysis and (b) total

plasmid recovery from bacterial lysates at the aqueous outlet.

The supercoiled plasmid form was clearly identified on the gel and seems to migrate

through the gel close to the velocity of a 5 kb linear fragment. Because the plasmid is

circular it will migrate through the gel at a rate which can be different from a linear DNA

(a)

(b)

Bacterial

Chromosomal

DNA

Plasmid

DNA

10 kb

8 kb

6 kb

5 kb

4 kb

3 kb

Ladder Microfluidic Chip Qiagen Kit Phenol Standard

In Out In Out In Out

44

fragment. Intact circular (supercoiled) plasmids will migrate through the gel faster than

their linear counterparts.

Based on band intensity, the pGLO concentration was estimated at 0.56 ng/l. Bacterial

chromosomal DNA was also found in the recovered solutions in the case of microfluidic

extraction and standard phenol extraction. In addition to this, a higher background

intensity can be seen in all of the sample inlet lanes on the gel for each of the three

purification methods. The recovered sample lanes have much lower background intensity

indicative of the successful removal of contaminants from the lysates using all three

extraction methods.

The efficacy of the plasmid extraction was compared with those obtained with a

commercial DNA kit (Qiagen) and standard phenol extraction. Using the same volume of

bacterial lysates, the plasmid recoveries using the microfluidic device, a Qiagen SPE

extraction kit and traditional phenol extraction, were 92.4 ± 8.4%, 95.1 ±2.5%, 98.3 ±

2.08%, respectively (Figure 3.8b). Therefore, the microfluidic phenol extraction shows

high recovery of DNA which is similar to efficiencies obtained with traditional extraction

methods.

In addition, the DNA extraction efficiencies found via microfluidic phenol extraction are

also comparable to the other microfluidic approaches using solid-phase extraction. Wen

et al work provides a comparison table of the solid-phase extraction device where the

DNA extraction efficiencies found to vary from 30% to 85% depending of the phase

material and the sample [56]. In comparison, the method proposed here would recover a

larger mass of DNA due to the lack of attachment and detachment steps present in the

solid-phase methods.

45

On the other hand, the protein quantification was not possible to achieve with the

microfluidic phenol extraction due to the phenol contamination found in the aqueous

phase. The phenol trace has a very high light absorption which impedes the DNA purity

analysis by absorbance ratios of 260nm/230nm and 260nm/280nm for organic

contaminants and protein content, respectively. Additionally, the bicinchoninic acid

(BCA) protein assay was also affected by the organic contamination.

The presence of a phenol trace in the aqueous phase is the main challenge of this work in

order to make this device suitable for integration with downstream processes required for

DNA manipulation as restriction cutting and PCR. In Chapter 6, some approaches to

address this issue are discussed.

Summarizing the first part of this work, microfluidic platforms were designed to perform

miniaturized liquid-phase extraction and to evaluate DNA purification and protein

partitioning efficiencies as well as plasmid extraction from bacterial lysates. Dual inlet-

dual output serpentine devices were utilized as prototype for the initial studies of protein

partitioning using stratified flow and droplet-based flow. Experiments infusing

fluorescently labeled proteins and DNA demonstrated the ability of the device to partition

proteins from the aqueous phase into the phenol phase and allowed a high percentage

DNA recovery in the aqueous phase with minimal DNA loss into the organic phase. A

redesigned device was fabricated to conduct DNA extraction from bacterial cell lysates

using only droplet-based flow, due to the mass transfer enhancement enabled by this

flow. Finally, plasmid extraction from transformed bacteria was conducted using this

microfluidic platform with DNA recoveries comparable to the recovery obtained by

standard macroscale extraction methods.

46

In Chapters 4 and 5 the electrokinetic manipulation of DNA is addressed as potential

continuation after the purification module presented here, as a pre-conditioning step prior

to other downstream analysis processes. Conclusions of this purification module and the

future work needed after this experimentation are discussed in Chapter 6.

47

Chapter 4

Background of Electrokinetic Sample Concentration

Most surfaces in contact with an electrolyte solution acquire a net charge due to various

charging mechanisms. As a result, these charged surfaces attract opposite-charge ions

(counterions) in the solution while repelling equal-charge ions (co-ions) leading to the

formation of an electric double layer (EDL) at the surface [57]. The thickness of this

double layer is commonly referred to as the Debye length λD. Electrokinetic phenomena

is a term applied to four distinct phenomena (electroosmosis, electrophoresis, streaming

potential, and sedimentation potential) that arise when the mobile region of the EDL and

an external electric field interact with a viscous shear layer near the charged surfaces

[58].

In this work, two of these phenomena, electroosmosis and electrophoresis, are considered

within a confined geometry by the application of a transverse electric field across the

width of a microchannel. A theoretical description of these phenomena is provided next

with the inclusion of the dielectrophoresis definition which will be later on required in the

discussion of the experimental results of a sample pre-conditioning module in Chapter 5.

In addition to the theoretical background, an up-to-date review of on-chip sample pre-

concentration methods is included here.

48

4.1 Electroosmosis

An electroosmotic flow is created by a net migration of ions of the EDL at a solid

charged surface due to the application of an external electric field as can be observed in

the right side of Figure 4.1. This ionic movement is transmitted to the bulk fluid due to

viscous forces causing a net fluid flow (slip velocity) relative to the stationary surface

[57, 58].

Figure 4.1: Schematic of the EP particle motion and the EO flow in an open

channel.

In order to obtain the electroosmotic velocity, the equations of conservation of

momentum (Navier-Stokes equations) need to be solved incorporating the electric body

force per unit volume given by E Ef E , where ρE is the charge density. Eq. 2.3 can be

rewritten as follows:

- +

+ - +

+ -

- -

- - + -

+ + - + -

- + +

- + - + + + -

- + -

-

Cathode Anode

- -

-

-

+

+

+

+

E

- - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - -

+ + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + +

UEO

UEP

λD

- - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + - + + + - + + +

+ - + - + - + + + - - + -

+ - - + -

- + -

λD

UEO

(b) (c) (a)

y

y

y

z

μEP ≈ μEO μEP > μEO

μEP=0

UEO UEO UEO

UP UP

UEO

UEP UEP

= =

+ +

49

(4.1)

If the flow is considered to be an inertia free capillary flow with no pressure gradient,

then Eq. 4.1 reduces to:

(4.2)

On the other hand, if we also consider Poisson‟s equation that relates the spatial

distribution of the electric field ( E ) to the charge distribution:

(4.3)

where ε is the permittivity of the solution and then plugging Eq. 4.3 in 4.2 we obtain:

(4.4)

Considering the reference system showed in Figure 4.1, the electric field has only

component in the z direction; therefore the flow velocity resulting is in that direction, uz.

In addition, the potential and uz are only function of y and their derivatives with respect z

are zero. Now, Eq. 4.4 can be reduced to the one-dimensional form as follows:

(4.5)

Integrating Eq. 4.5 gives

(4.6)

by setting at the bulk of the fluid (y→∞) ∂uz/∂y and ∂ϕ/∂y equal to zero. Integrating

again and setting at the hydrodynamic slip plane (y=0) uz=0 and ϕ= ζw. This then gives

wEO zU E

(4.7)

2

Ept

uu u u E

2 - E u = E

2 E

2 2 u = E

2 2

2 2

zx

uE

y y

=

zz

uE

y y

=

50

This is the electroosmotic velocity developed by the bulk flow and it is called the

Smoluchowski slip velocity equation [58]. The electroosmotic (EO) mobility is then

defined as

wEO

(4.8)

Summarizing, when applied axially along a channel and under constant longitudinal

electric field, the electroosmotic velocity developed by the bulk flow is proportional to

the electric field and it is equal to the Smoluchowski slip velocity.

The zeta potential of the wall determines the magnitude of the EO flow. The wall zeta

potential is affected by ionic strength, counterion type and pH of the solution [59-63], but

is also dependant on the surface properties of the wall. The effect of channel material

and/or surface modifications to the development of the EDL has been widely investigated

to enhance or suppress EO flow [64-67].

Nevertheless, the most important use of the EO flow is to drive bulk fluids in

microchannels and it provides the advantage over other methods that the electrodes can

be embedded within the chip allowing compact microchips and precise fluid control [32,

68].

4.2 Electrophoresis

Electrophoresis is defined as the motion of a charged particle through an electrolyte

solution under an applied external electric field [58]. The electrophoretic force induced

by the external field is given by

51

EPF qE (4.9)

where q is the net charge between the charge particle and the concentric Debye length

(λD) as depicted in Figure 4.1 (left side). Considering a large Debye length, where λD is

larger compared with the analyte radius a and balancing the electrical force with the

Stokes drag force, it gives:

6 EPqE U a (4.10)

From the definition of surface charge, the potential can be written as [58]:

4 ( )P

D

q

a

(4.11)

where ζp is the zeta potential of the particle in that solution. Now the particle velocity

with a large Debye length results in:

(1 / )2 2

3 3

P D PEP

a E EU

(4.12)

However, the migration velocity due to the electrophoretic effect for a thin diffuse EDL

at the particle surface can be reduced to the Helmholtz-Smoluchowski equation [58]:

PEPU E

(4.13)

The electrophoretic (EP) mobility is similarly defined as

PEP

(4.14)

It should be noted that the EP mobility for thin EDL is not dependent of the shape and

size of the particle. The size effect is implicitly contained in the zeta potential. The zeta

52

potential of the particle, directly affecting the particle mobility, is a function of the net

charge resulting of the charged particle exposed to the electrolyte solution. However,

larger DNA fragments are free draining so the free solution mobility has been found to be

independent of the molecule size, and size fractionation can only be produced by physical

interaction of DNA within a confining geometry like capillaries, sieving gels or

nanostructures where fragment motion is retarded based on their size [69]. Stellwagen et.

al. measured DNA free solution mobility for various sized DNA fragments in the most

commonly used electrophoretic buffers, demonstrating size independence on DNA free

solution mobility for fragments lengths greater than 400 bp [67].

It has also been shown that the DNA EP mobility decreases with increasing buffer ionic

strength and conductivity due to the compression of the EDL and the reduction of the

zeta potential. Lastly, counterion type can affect DNA mobility, where the DNA mobility

is significantly reduced in the presence of high NaCl concentration buffers, such as

phosphate-buffered saline (PBS), due to electrostatic shielding of surface charges on

DNA phosphodiester backbone [70, 71].

4.3 Combined electrophoresis and electroosmosis in confining

geometries

As a result of EP and EO phenomena, charged analytes will move at a velocity that

depends on both the EP mobility of the analyte and any induced EO flow. The net

velocity of a particle is therefore a superposition of the EP velocity and EO velocity due

53

to drag on the particle from the EO flow. In a unidirectional linear flow field the net

velocity of a charged particle is

EUUU EOEPEOEPP (4.15)

Thus, the velocity of a DNA or protein molecule will be not only affected by its net

charge but also by the flow convection due to the EO flow as shown in Figure 4.2.

Figure 4.2: Net particle velocity Up as a result of the EP force and the EO flow drag

force.

If the driving electric field is applied across the width of a microchannel, the

electroosmotic velocity at the top and bottom of the channel is predicted by Eqn. 4.7.

However, since the channel has bounding walls the flow must recirculate in order to

conserve fluid mass similar to the classic lid driven (Couette) cavity flow problem.

In this situation, the top and bottom walls move at the slip velocity (UEO) depending on

their wall zeta potentials. This original movement imposes a constant pressure gradient

for the conservation of mass which leads to the one-dimensional Navier-Stokes equation:

2

2

12zu P

ay z

(4.16)

Integrating Eq. 4.16 twice gives:

2

zu ay by c (4.17)

PEPU E

w

EOU E

P EP EOU U U

54

Let consider now the following boundary conditions:

2 0bottomz EOu ay by c U at y (4.18)

2

topz EOu ay by c U at y H (4.19)

where H is the thickness of the channel. Finally, the mass conservation condition needs to

be satisfied:

2 3 2

0 00 0 0

3 2

H H

z

a bu dy ay by c dy H H cH (4.20)

Therefore, from Eq 4.18, when y=0, c can be obtained, c=UEObottom. Then, a and b can be

found using equations 4.19 and 4.20 resulting in:

2

3top bottomEO EOa U U

H (4.21)

22

top bottomEO EOb U UH

(4.22)

The recirculation profile is shown schematically in Figure 4.3a with the velocity profile

as a function of channel depth at the center of the channel. The net flow rate of the

recirculation is zero due to conservation of mass. An uncharged particle (EP=0) would

follow the streamlines set up by this flow profile.

Such transverse electroosmotic flow profiles have previously investigated but for

micromixing enhancement purposes to actively mix two solutions [72]. These flows can

be divided into DC electroosmosis (time-independent), where non-uniform zeta potentials

are used to induce a change in the surface charge distribution to mix fluids,[72-74] and

AC electroosmosis (time-dependent), where high frequency (~1 khz) and low frequency

55

(~1 hz) EO flow recirculation has been studied [75-78]. In addition, by using spatially

patterned electrodes integrated along the length of a microchannel, a helical recirculation

path down the length of the channel was produced which promoted mixing of rhodamine

B dye using DC fields [79, 80].

Figure 4.3: EP velocity, transverse EO flow velocity and net particle velocity in a

closed glass/PDMS microchannel (cross-sectional view). (a) EO flow only (μEP=0),

(b) EO flow and high EP force (μEP>μEO), and (c) EO flow and EP force with similar

mobility values (μEP≈μEO).

(b) (c) (a)

y

y

y

z

μEP ≈ μEO μEP > μEO

μEP=0

UEO UEO UEO

UP UP

UEO

UEP UEP

= =

+ +

56

However, if the particle has a significant surface charge, it will feel an additional EP

force where the local particle velocity is the superposition of the EP velocity with the EO

flow velocity. When the EP mobility is larger than the EO mobility (μEP > μEO) then the

particle velocity at the channel center is estimated look like Figure 4.3b and have a

positive magnitude throughout the depth of the channel. In this case, the particle can

migrate towards the biased electrode, traverse the channel width, and ultimately collect at

the electrode surface.

If the EP mobility and EO mobility are approximately of equal magnitude (EP EO)

then the particle velocity at the channel center will behave as shown in Figure 4.3c. In

this case, the wall velocity is now expected to be close to zero while the rest of the

velocity profile is in the positive direction. Any particle flowing with a net positive

velocity (away from the channel top and bottom) will ultimately recirculate around and

back across the channel due to the recirculation flow profile set up by the bounding wall.

This particle will ultimately migrate to an equilibrium position where the net velocity is

zero (either at the wall or in the center of a vortex), allowing trapping of the particle at

this position.

4.4 Dielectrophoresis

The dielectrophoretic phenomenon is defined as particle motion generated by an induced

dipole moment within the dielectric particle when the particle is exposed to a nonuniform

electric field. The DEP dielectrophoretic force does not require a net charge on the

particle but involves is brought about due to the relative difference in polarizability of the

57

particle and surrounding medium through the complex dielectric particle permittivity (εp)

and a dielectric medium permittivity (εm) to induce the particle dipole moment. The

effective dipole moment for a spherical particle with radius a is given by [68]

* *

3

* *4

2

p m

m

p m

p a E

(4.16)

The complex dielectric constant is defined as

*

j

where ε is the respective

dielectric constant, σ is the electrical conductivity, ω is the electric field frequency, and j

the imaginary number. The term in parenthesis is also known as the Clausius-Mossotti

factor (fCM). The DEP force can be described by

DEPF p E (4.17)

Then the effective DEP force is given by [68]

232 ReDEP m CMF a f E (4.18)

From the preceding equation it can be noticed that the DEP force is dependent on the

particle size and the electric field gradient squared. Additionally, the field polarity does

not affect the migration due to the square dependence and therefore, the force can be then

generated by either AC and DC fields.

Moreover, the Clausius-Mossotti factor fCM plays an important role in the DEP force,

giving the resulting sign of the force. This factor is frequency dependent then there is a

crossover frequency where the force changes sign. When real part of fCM is positive, the

particle moves toward areas of high field (positive DEP), typically electrode edges, while

when it is negative, particle moves away from high field regions (negative DEP). Due to

58

this frequency dependence of fCM, DEP can be modulated from positive to negative DEP

and vice versa simply modulating the driving frequency above the cross over frequency

specific to a particle type using applied AC fields. DEP has been used as an attractive

technique to concentrate or separate particles them from other analytes through

differences in relative polarizability.

4.5 State-of-the-art of sample concentration devices by electrokinetics

forces

To date, the concentration techniques utilized to concentrate both DNA and proteins in

diluted sample have included the use of electrophoretic and dielectrophoretic forces. The

electrophoretic methods are sample stacking techniques including field-amplified sample

stacking (FASS), isotachophoresis (ITP), and pH-mediated stacking. FASS is one of the

most widely used pre-concentration methods for both DNA and proteins because it is one

of the simplest techniques; achieved by preparing the sample solution with conductivity

lower than that of a background electrolyte. An abrupt change in conductivity between

the sample and background electrolyte from low conductivity to high conductivity causes

a rapid reduction of the local electric field which slows the electrophoretic velocity of an

analyte as it passes into the high conductivity region. This velocity reduction ultimately

produces a local accumulation of the analyte at the conductivity interface (Figure 4.4).

Theoretical and experimental studies were presented first by Burgi and Chien [81] and

then Santiago‟s group provided a mathematical model of the sample concentration [82].

Moreover, FASS has been investigated in combination with CE injection strategies

59

improving up to 160-fold the original signal [83]. Also, the use of porous polymer

structures to enable a pressure driven injection scheme while avoiding instabilities due to

high-conductivity gradients demonstrated a signal increase by a factor of 1100 in

fluorescein and Bodipy electrophoretic separations [84].

Figure 4.4: FASS stacking: (a) schematic of the low-conductivity sample loading into

a high-conductivity buffer and (b) epi-fluorescence images during the stacking [82].

In ITP, a multi-analyte sample is sandwiched between two electrolytes, termed leading

and terminating electrolytes, with faster and slower electrophoretic mobilities than the

sample solution, respectively. Analytes move forward during voltage application forming

discrete high concentration zones according to their electrophoretic mobilities. After a

transient period where the discrete zones are formed, the samples stack when moving at

constant velocity in the direction of the leader as it is shown in Figure 4.5. ITP has been

reported for on-line sample preparation to enrich analytes with low initial concentrations

with enhancements in range from 40 fold for SDS-protein [85], 400 folds in ITP-zone

electrophoresis chip [86] and up to 100 000 folds for small ions [87]. ITP is not

commonly used for the separation of nucleic acids, but rather as a pre-concentration step,

60

since DNA free solution mobility is relativity constant over a large range of DNA sizes

(>0.4 kb) [67].

Figure 4.5: Images of isotachophoresis stacking of two samples with different

electrophoretic mobility [88].

pH-mediated stacking is another method to pre-concentrate samples by using a pH

gradient to decelerate and stack samples (Figure 4.6) similar to the FASS approach

described before. A plug containing a sample in a low pH matrix is injected into a

microchannel previously filled with a high pH background electrolyte.

Figure 4.6: Schematic of pH-mediated stacking methodology where analytes migrate

to the isoelectric point [5].

61

A steep pH boundary is developed at the front end of the plug where cationic analytes

dissolved in the low pH plug are made anionic once they enter the high pH background

electrolyte. This effect focuses the analytes at the pH interface by retarding their

movement as they migrate to a detection zone [89]. This approach can not only be used

as a pre-concentration method but also as an analyte separation technique using

differences in isoelectric point [5, 90]. pH mediated stacking is most applicable to protein

solutions due to their widely varying isoelectric points. DNA mobility, on the other hand,

is fairly constant over a wide pH range [91].

In addition to purely electrophoretic methods for sample concentration, DEP has been

investigated as an additional electrokinetic technique used for sample trapping. As

previously noted, when exposed to a non-homogenous electric field, polarizable particles

including cells, proteins and DNA can develop a strong electric dipole moment

depending upon the relative polarizability of the particle with respect to the surrounding

electrolyte that can be used to move them toward (positive DEP) or away (negative DEP)

from areas of high electric fields in order to concentrate or separate them from other

particles.

Since the dielectrophoretic force is dependent on the square of the field gradient, it often

employs alternating current (AC) fields in order to avoid any additional electrophoretic

effects, with the polarizability of the particle (positive or negative DEP) being AC

frequency dependant as explained previously. Trapping of DNA has been investigated by

using DEP with AC electric fields [92, 93] as can be seen in Figure 4.7. In addition, DEP

was used in combination with electrode geometric configurations which create non-

uniform fields [94, 95] as well as with AC electroosmotic flows [96].

62

Figure 4.7. Trapping of DNA molecules at the edges of gold-film strips using

dielectrophoresis. (a) and (c) before field application, (b) and (d) 1 min after

applying a 30-Hz, 200-V p-p signal, nearly all of the molecules became trapped [93].

Additionally, other methods have been explored to capture and concentrate samples using

chemical binding through the use of streptavidin-coated magnetic beads,[97]

photoactivated polycarbonate surfaces [18] for DNA concentration, and the utilization of

membranes for protein concentration enhancements [98, 99]. A size exclusion membrane

and the chip schematic used for protein concentration amplification are shown in Figure

4.8.

63

Figure 4.8: Chip layout and image of a photopolymerized size exclusion membrane

where analytes are trapped and concentrated [98].

In next chapter, a novel concentration method is demonstrated to continuously enhance

DNA and protein sample signal by balancing the electrophoretic force with a transverse

electroosmotic flow employing uniform electric fields applied across the width of a

microchannel.

64

Chapter 5

Sample Trapping and Concentration

In this chapter, a converging Y-inlet microfluidic channel with integrated coplanar

electrodes is described and tested using transverse uniform DC and AC electric fields to

assess the ability of DNA and protein to traverse and concentrate as a preconditioning

step for downstream processes.

5.1 Device Fabrication

The microfluidic channel was designed as a two inlet converging and two outlet

diverging geometry with a daughter channel 20 μm in depth, 300 μm wide and 5 cm long.

It was fabricated using photolithography and soft lithography techniques as previously

described in the Chapter 3. The master mold was lithographically patterned using SU-8

negative photoresist (MicroChem, Newton, MA) and then cast in polydimethylsiloxane

(PDMS) by mixing the elastomer base and curing agent (Sylgard 184 Silicone Elastomer

Kit, Dow Corning, Midland, MI) at a ratio of 10:1 and poured on top of the master mold.

The PDMS was degassed in a vacuum chamber and then cured in an oven at 65˚C for 1

hour.

65

Figure 5.1: Schematic of the fabrication steps required to obtain metal electrodes on

glass substrate.

On the other hand, the electrode design was a pair of co-planar electrodes with a center to

center separation distance of 231 μm and electrode width and length of 45 μm and 5 cm,

respectively. The fabrication was made by patterning the feature from a mask onto a glass

slide using a lift-off technique as described in Figure 5.1. First, the slide was

lithographically patterned and developed using Microposit S1818 positive photoresist

(Microposit, Marlborough, MA, USA). In the exposed area, the positive photoresist

washed away during developing leaving the glass exposed for deposition. Then, the

patterned glass slides were recessed slightly by submerging the slides in 5:1 buffered

66

hydrofluoric acid for 1 minute. Subsequently, the etched slides were placed in a metal

sputtering system (Kurt Lesker PVD 75, Kurt J. Lesker Company, Pittsburgh, PA) to

deposit titanium and platinum resulting in a total electrode thickness of 290 nm. Finally,

the photoresist was removed by dissolution in acetone bath using a sonicator. An

electrical connection was made by depositing CW2400 conductive epoxy (Chemtronics,

Kennesaw, GA) onto the terminals of the patterned electrodes and attaching wires

connected to an external power supply. The electrode fabrication steps and settings are

also described in more detail in Appendix A.

Figure 5.2: (a) Schematic of the electrical and microfluidic components of the device

and (b) photograph of the PDMS microchannel with co-planar electrodes.

67

The glass slide containing the electrode pattern was then irreversibly bonded to the

PDMS microchannel structure by surface activation of the glass and PDMS with oxygen

plasma at 100 W, 300 mTorr, for 60 sec. Small drops of methanol were placed on both

the glass and PDMS surfaces after the surface modification to provide lubrication during

the alignment process. The two components were aligned under a microscope and

brought into contact to initiate the bonding process. The bonded device was cured at

65°C for one hour. Figure 5.2 shows a schematic of the two device components (Fig

5.2a), PDMS microfluidic channel and the patterned electrodes on the glass slide, as well

as a photograph of the assembled device with both electrical and fluid connections (Fig

5.2b).

5.2 Materials and Buffer Characterization

DNA and protein samples of known concentrations were infused into the device diluted

in three different buffers to investigate their influence in the DNA migration. The buffer

was either 1× TBE buffer (89 mM Tris Base, 89 mM Boric Acid, 2 mM EDTA pH 8.0)

(Rockland, ME), 1× TE buffer (10 mM Tris-HCl, 1 mM EDTA pH 8.0) (Rockland, ME),

or 1× PBS (HyClone, Thermo Scientific, Logan, UT). Lambda phage DNA (48 kbp) and

pUC19 vector (2,686 bp) were purchased from New England BioLabs (Beverly, MA) and

labelled using a fluorescent YOYO-1 intercalating dye as described in Chapter 3. All

DNA samples were prepared at a 0.05 μg/μl as final concentration. BSA protein

(Omnipur, fraction V; EMD Biosciences) was conjugated with 5-6-carboxy-tetra-

methylrhodamine (TAMRA) succinimidyl ester dye (Fluka, Switzerland), using the same

labeling procedure explained in Chapter 3, and finally diluted at a 0.25 μg/μl

68

concentration in the appropriate buffer solutions (1× TE buffer and 1× TBE buffer). The

sample preparation protocol is described in Appendix B.

The buffer characterization was also made to obtain electrical properties of the buffers

used in this work. That includes electrical conductivity, ionic strength, zeta potential and

electroosmotic velocity, summarized in Table 5.1.

The electrical conductivity is the buffer's ability to conduct an electric current and can be

calculated as

2 2

1

nF c zi i i

i

, where F is the Faraday constant, ci is the molar

concentration, vi the ionic mobility and zi the charge of species i [58]. The total buffer

conductivity was measuring using a digital conductivity meter (Oakton Instruments,

Vernon Hills, IL, USA). For each buffer, three different readings were taken and then

averaged, with prior rinsing of meter probe in DI water between the buffer measurements

to assure a correct reading.

Table 5.1. Comparison of electroosmotic mobilities for buffer used

The ionic strength is the measurement of the ion concentration in that solution which

plays a central role to describe the formation of the electrical double layer at the wall. For

Solution Conductivity

(S/cm)

Ionic Strength

(mM)

Zeta Potential

(mV)

EO Mobility

(cm2/Vs)

TBE 859 105 -48.05 3.40×10-4

TE 739 13 -68.13 4.80×10-4

PBS 15000 137 -22 1.50×10-4

69

each buffer, the ionic strength was calculated using the expression

1 2

12

nI c zi i

i

[100]. Finally, the zeta potential resulting in glass/PDMS microchannel walls wetted with

the buffers used in here were found in literature [101] and used to calculate the

electroosmotic mobility by employing Eq. 4.5, where ε= εoεr and considering water

properties ( εr = 80 and μ = 0.001 Pa·s).

5.3 Experimental Setup

The microfluidic device was connected with 0.254 mm ID Tygon tubing (Small Parts

Inc., Miami Lakes, FL) to glass syringes (Hamilton, Reno, NV) and the fluid infusion

was made via syringe pumps (PicoPlus 22, Harvard apparatus, Holliston, MA).

Since the center to center electrode spacing is 231 m, a relatively large electric field can

be generated at low DC voltages (3 VDC or less) in order to avoid solution electrolysis.

The electric field was applied as a constant DC field using a DC power supply (Agilent

Technologies, model E3631A). A pulsed DC and AC fields were also be employed by

using a signal generator (Agilent Technologies, model 33220A) and high-voltage

amplifier (Tegam model 2350, Tegam Inc., Madison, OH).

In order to visualize the YOYO-1 labelled DNA the fluorescence intensity in the

microchannel was observed using a Nikon Eclipse TE2000U inverted microscope

(Nikon, Tokyo, Japan) operating in epifluorescence mode with a FITC filter cube

(Chroma Technology, Brattleboro, VT) and a charged coupled device (CCD) camera

(PowerView 1.4 MP, TSI Incorporated, Shoreview, MN) was used to acquire microscopy

70

images. All the images were acquired using a 10× objective, neutral density filter ND1

for the epifluorescence arc lamp (Nikon, Tokyo, Japan), a 100-ms exposure time at 5

frames per second. The intensity profiles of the images were determined using NIH-

ImageJ software (National Institutes of Health), employing plot profile function, saving

the XY values and plotting them in Matlab (Mathwork Inc.) for better visualization.

Concentration enhancement was determined by pixel intensity analysis of the acquired

epifluorescent images. These images were used to obtain intensity profiles after

background elimination. Concentration enhancement was calculated by normalizing the

peak fluorescence intensity of the concentrated DNA after field application to the mean

intensity of the solution prior to the application of the electric field. Mass conservation

was estimated by using the intensity integral of the DNA concentration profile by the sum

of the pixel intensity values above a set threshold across the image width.

5.4 Results and Discussion

Transverse migration of DNA in 1× TBE

In order to investigate the DNA migration within a transversely applied electric field,

initially DNA samples diluted in 1× TBE buffer were pumped at a fixed flow rate (0.5

l/min) conjunctly with a clean 1× TBE buffer solution into the main channel where a

DC voltage was applied transversally across the channel via the coplanar electrode pair.

71

Figure 5.3: Transverse electrophoretic migration of pUC19 vector in 1× TBE buffer

(bottom-up view). (a) No field is applied, (b) when 1 VDC was applied, (c) 2 VDC and

(d) 3 VDC.

Pictures of pUC19 samples at the outlet of the device under different field conditions are

shown in Figure 5.3. When a 1 VDC voltage (Eave=43.29 V∙cm-1

) was applied (Figure

5.3b) DNA migration was not observed and the infused fluids moved parallel to each

other just as when no field was applied (Figure 5.3a). When the voltage was increased to

2 VDC (Eave=86.58 V∙cm-1

), the electrophoretic force was sufficient to produce transverse

DNA migration toward the positive electrode located at the bottom of the images.

However, the DNA residence time in the microchannel was not long enough for the DNA

to traverse the entire width of the channel (Figure 5.3c). When the voltage was increased

to 3 VDC (Eave=129.87 V∙cm-1

), the DNA EP velocity was then high enough that the DNA

72

could accumulate at the positive electrode and was convected to the device outlet by the

axial pressure-driven flow (Figure 5.3d). The DNA would concentrate at the electrode in

approximately 15 seconds after turning on the power supply.

Figure 5.4: Fluorescence intensity profiles of pUC19 (a), and -phage DNA samples

(b) in 1× TBE buffer across the channel width and at different times after the field is

applied showing the transverse DNA migration.

In addition to the experiments using pUC19, circular double-stranded DNA plasmid

(2,686 bp), same experimental conditions were repeated using λ-phage DNA, linear

73

double-stranded linear DNA (48 kbp), observing the same electrophoretic movement

towards the positive electrode. Intensity profiles for both types of DNA at the outlet of

the device as a function of time are shown in Figure 5.4. DNA was found to be increased

4 fold for pUC19 and 2 fold for -phage DNA over the original infused DNA intensity

according to the peak fluorescence intensity.

The area under the fluorescence intensity profile was also calculated, comparing the

intensity integral between the initial (no field) and final (concentrated) DNA solutions to

estimate the total amount of DNA in the images. The concentrated intensity areas were

169% and 110% compared to the original intensity profile without an applied field, for

pUC19 and -phage DNA respectively.

No significant difference was found in the temporal evolution of the DNA trapping at the

positive electrode between the short (pUC19) or long (-phage) DNA types, agreeing

with previously observed results demonstrating size independence during free solution

electrophoresis for DNA fragments bigger than 400 bp [67].

Transverse migration of DNA in 1× TE

Following, the same experimental conditions were repeated except that the solution

buffer was changed, replacing the TBE buffer with 1× TE buffer in both the DNA

solution and the receiving buffer. By changing the electrolyte solution to a low ionic

strength buffer, the DNA then showed a different migration behavior (Figure 5.5)

compared to when 1× TBE was used as a buffer (Figure 5.3).

74

Figure 5.5: pUC19 electrophoretic migration and concentration in 1× TE buffer

(bottom-up view). (a) No field is applied, (b) when 1 VDC was applied, (c) 2 VDC and

(d) 3 VDC.

The DNA solution in TE buffer using different voltages values are depicted in Figure 5.5.

As shown in Figure 5.5b-c, the lower applied voltages produce very little DNA migration

across the channel width. However, when the voltage is raised to 3 VDC (Figure 5.5d), the

DNA migrates towards the positive electrode but it reached a stable position in the center

of the channel where it was trapped and concentrated. This is in contrast to Figure 5.3d

where the DNA migrates completely across the channel to the positive electrode.

Following the application of voltage, it takes the DNA approximately 10 seconds to

migrate to its trapped position.

75

Figure 5.6: Fluorescence intensity profiles of pUC19 (a) and -phage DNA samples

(b) in 1× TE buffer across the channel width and at different times after the field is

applied showing DNA trapping and concentration at the channel center.

Figure 5.6 compares the temporal intensity profiles for both pUC19 and λ-phageDNA as

they migrate to their steady state position in the center of the channel, again independent

of the DNA size and basepair number as observed in Figure 5.4. The peak fluorescence

intensity of the concentrated DNA increased 2 fold for pUC19 and 3 fold for -phage

DNA over the infused DNA intensity. In addition, the concentrated intensity areas were

76

73% and 129% compared to the zero field intensity profile, for pUC19 and -phage DNA

respectively.

For both experiments using TBE and TE buffers, the trapping location and transient

behavior were highly reproducible over multiple experiments (n=3 for TBE, n=5 for TE).

The absolute concentration enhancement varied between experiments, but based on the

fluorescence intensity profile of the DNA after it had concentrated at its equilibrium

position in both TBE and TE buffers, the DNA concentration clearly increased several

fold over the initial infused concentration.

When analyzing fluorescence intensity profiles, it was found that the concentrated

intensity integrals which are predictive of DNA mass were usually within ± 30%

compared to the zero field intensity profile. Errors in mass quantification are attributed to

two main factors, fluorescence quantification and non-specific DNA absorption onto the

microchannel surfaces. Since the DNA concentrates to a thin line, the concentrated

intensity integrals are only summed over a pixel width of ~50 pixels which can lead to

errors in the integral quantification. Second, due to non-specific absorption of DNA to

the channel surfaces DNA could be lost in other areas of the channel or accumulate at the

location where the DNA concentrated leading to a larger measured fluorescence intensity.

This was particularly apparent when using a low ionic concentration buffer such as TE,

where a non-specific background fluorescence intensity can be seen in the top half of the

channel even after the DNA was concentrated (Figure 5.5d). It was also found that when

channels were flushed with buffer following DNA concentration, the area where the

DNA was concentrated was still slightly fluorescent indicating non-specific adsorption of

the DNA to the channel surface (data not shown).

77

Transverse migration of proteins

In addition to experiments using DNA samples, experiments were conducted to assess the

ability to trap and concentrate proteins. When using anionic BSA protein (pI=4.7), the

same trapping behavior seen when using DNA in both TE and TBE buffers was observed

(Figure 5.7). When TBE was used as a buffer, the protein migrated completely to the

positive electrode (Figure 5.7b) but was trapped and concentrated in the channel center

when diluted in TE buffer (Figure 5.7c), similar to the DNA behavior in Figure 5.3 and

5.5, respectively.

Figure 5.7: Rhodamine labeled BSA migration using 1× TBE buffer and 1× TE

buffer. (a) image of BSA when no field was applied, (b) BSA in 1× TBE buffer when

3 VDC is applied, (c) BSA in 1× TE buffer when 3 VDC is applied, and (d)

fluorescence profiles across the channel width.

78

The intensity profiles are shown in Figure 5.7d and appear to be slightly wider than those

observed when using DNA. The possibility of concentrating proteins gives this method a

new prospective to assess lower detection points for multiple diagnostic applications.

Further investigation of migration dynamics using pure AC field and pulsed field

In experiments using either TBE or TE buffers, the electrical conductivity across the

width of the channel is homogeneous, so the electric field should also be constant across

the width of the device so other electrokinetic phenomena like FASS, isotachophoresis or

dielectrophoresis are not expected.

Several experiments were conducted to better understand the dynamics of DNA

migration observed in order to rule out other mechanisms such as DEP, axial flow

dependence, to enhance the migration, and to observe the recirculation flow profile

predicted in Figure 4.3a.

First, a pure AC field or square wave was applied across the electrodes to investigate the

occurrence of dielectrophoresis effects due to field inhomogeneities. Figure 5.8 shows the

DNA sample diluted in 1× TE buffer and infused at 0.5 μl/min during the application of

an square wave (3Vpeak at 1 MHz). As a result, the DNA did not migrate transversely

across the channel as seen when using DC electric fields. When using AC or square wave

fields, the time averaged applied voltage is zero so analytes move in one direction during

the first half cycle of the wave and in the opposite direction during the other half cycle,

resulting no noticeable net migration. It is noted that the trapping observed using DC

fields is the result of the field polarity. In addition, this demonstrates that DNA

79

dielectrophoresis, which is polarity independent, is not the trapping mechanism seen in

this work.

Figure 5.8: DNA sample under alternative square wave 3 Vpeak-to-peak (bottom-up

view).

To observe any influence of the axial parabolic flow on the DNA migration, the perfusion

flow rate was also varied. In Figure 5.9, images of pUC19 in 1× TE buffer were acquired

at the entrance of the channel, where three flow rates were compared: 0.5 μl/min (5.9b),

0.3 μl/min (5.9c) and 0.1 μl/min (5.9d). The flow rate affects the position along the length

of the channel where the DNA concentrates but not the transverse location and width of

the concentrated DNA plug. A decrease in the axial velocity allows the DNA to

concentrate closer to the entrance of the channel, since the DNA is convected forward at

a slower velocity. At 0.5 μl/min, the initial DNA migration can be seen but DNA trapping

occurs further down the channel and was not observed at entrance region. At 0.3 μl/min

and 0.1 μl/min, the DNA begins to concentrate towards the center of the channel and is

captured in the images at the channel entrance. As shown here, the pressure driven flow

only affects in the residence time of the trapping but the transverse location and width of

the concentrated DNA plug did not change.

80

Figure 5.9: Images at the main channel entrance of DNA samples in 1× TE buffer at

different flow rates (bottom-up view). (a) No field, 0.5 μl/min flow rate, (b-d) 3 VDC

was applied when the flow rate was (b) 0.5 μl/min, (c) 0.3 μl/min, and (d) 0.1 μl/min.

The voltage used to promote DNA migration was limited by buffer hydrolysis. At applied

voltages larger than 3 VDC, formation of electrolysis bubbles was apparent. Experiments

were also conducted using a pulsed DC field which showed identical DNA trapping and

concentration at the same channels positions as a continuous DC field (Figure 5.10). The

pulsed electric field consists of a pulse signal of 15 VDC at a 1 MHz frequency and 20%

duty cycle (so the pulse is only applied for 200 ns followed by a 800 ns rest period).

81

Figure 5.10: Images of pUC19 sample in TE buffer as time increases under a pulsed

electric field of 15 VDC at a 1 MHz frequency and 20% duty cycle (bottom-up view).

In this manner the DC field is not applied long enough at each pulse for electrolysis

bubbles to nucleate. This allows a higher total electric field to be applied without solution

electrolysis. The same DNA concentration behavior is seen as when using a constant

electric field, but the DNA accumulates at its equilibrium position much faster that the

DC case, reaching the steady state in approximately 5 sec.

Effect of EO and EP mobility on DNA migration and concentration

After discharging the dielectrophoresis force as the driving force for the trapping seen in

TE buffer, in this section the other two electrokinetic phenomena, EO flow and EP force

are analyzed to explain the mechanism which trap and concentrate DNA and proteins in

the center of the channel or accumulate them in the positive electrode, depending of the

buffer chosen.

As noted previously, the buffer ionic strength plays an important role in the zeta potential

formation. As the ionic strength increases, the double layer becomes compressed

resulting in a lower zeta potential and hence lower electroosmotic mobility [61-63].

Therefore, the electroosmotic velocity will be faster in low ionic strength buffers like TE

82

buffer and slower in TBE and PBS buffers. The EO mobilities are calculated to be

4.80×10-4

, 3.40×10-4

and 1.50×10-4

cm2∙V

-1∙s

-1 in TE, TBE and PBS respectively (Table

5.1).

The free-solution EP mobility of DNA is also buffer dependent, decreasing

monotonically as the buffer conductivity and salt concentration increase [70, 71]. From

literature, the DNA electrophoretic mobility in TBE was found to be 4.5 ×10-4

cm2∙V

-1∙s

-1

[67]. No report was found regarding DNA mobility in PBS or TE buffer. However, based

on the conductivities of these buffers (Table 5.1), the DNA mobility in TE is expected to

be slightly higher that the TBE mobility while the DNA mobility in PBS is expected to be

much smaller due to the high NaCl salt concentration (137 mM) and electrostatic

shielding [70, 71].

The observed migration behavior of DNA in different buffers could be explained by a

balance of EP and EO forces and the effect of the different buffers on the relative EP and

EO mobility, as illustrated in Figure 4.3. By changing the solution buffer, the EP and EO

mobility can be parametrically changed to observe different migration dynamics and

equilibrium DNA positions. The effect of varying buffers is shown in Figure 5.11. The

images were acquired at the device outlet showing pUC19 concentration at different

locations along the width of the channel depending on the buffer conditions. When a

transverse electric field (3 VDC) and pressure-driven flow (0.5 μl/min) were applied,

different concentration dynamics were observed using three different buffers.

83

Figure 5.11: Effect of buffer selection on pUC19 migration under transverse DC

electric field. (a) No field applied, (b-d) 3 VDC was applied while pUC19 was diluted

in three different buffers: (b) 1× TBE, (c) 1× TE, (d) 1× PBS.

When TBE was used as a buffer, the EP mobility (4.50×10-4

cm2∙V

-1∙s

-1) is larger than the

EO mobility (3.40×10-4

cm2∙V

-1∙s

-1) so the net DNA velocity should be positive

everywhere along the depth of the channel as shown in Figure 4.3b allowing the DNA to

transverse the channel to the positive electrode (Figure 5.11b).

When TE was used as a buffer, the EP and EO mobility are approximately equally

(~4.50×10-4

versus 4.80×10-4

cm2∙V

-1∙s

-1) due to the low ionic strength of the TE buffer,

so the migration dynamics are expected to follow those shown in Figure 4.3c where there

is a zero velocity point (either at the wall or in the center of the recirculation vortex)

which traps the DNA at the center of the channel (Figure 5.11c).

84

Finally, when PBS was used as a buffer (Figure 5.11d) the DNA migrated partially

toward the positive electrode while a large portion of the sample remained in the left side

of the channel. The fact that the DNA could not transverse the channel is attributed to

both a decrease in the EP and EO mobility so that the recirculation vortex is less

prominent while the EP force is also reduced slowing the DNA migration across the

channel.

The balance of these two forces, electrophoretic force and the drag force induced by EO

flow, therefore determines the DNA migration across the channel. Next, two different

approaches are presented to demonstrate the formation of an induced helical EO flow by

using numerical simulations and particle visualization.

EO Flow Simulation

The physical model was simplified by assuming a two-dimensional problem using the

cross-section area along the channel width. The electrodes are located at the bottom

surrounded by glass and connected by top and side PDMS channels (Figure 5.12).

Figure 5.12: 2D geometry representing the cross-section area of the channel.

COMSOL Multiphysics (version 3.3, Burlington, MA) was used for these simulations

where the electroosmotic flow can be simulated with MEMS module, by selecting

Electroosmotic Flow case in the Microfluidics section and then, Stokes Flow and the

0 V +3 V

y y

z y

85

Steady-State Analysis. The Stokes Flow component is used to solve the Stokes equation

including the electroosmotic contribution and continuity equation:

2

Eu p V (5.1)

0u (5.2)

where μ is the fluid viscosity, u the velocity, p the pressure, ρE the charge density and V

the voltage. At the same time, the Conductive Media DC component is used to solve

Poisson equation:

2

o

V

(5.3)

being εo the vacuum permittivity. The details of all the COMSOL settings for these

simulations can be found in Appendix C.

Figure 5.13: Measurements of zeta potentials for selected buffers using

PDMS/PDMS and PDMS/glass channels [101].

86

In order to simulate the EO flow using COMSOL software, it is important to know the

zeta potential or EO mobility of the buffer used at the wall of interest, which could be

PDMS (top and side walls) or glass (bottom).

According to the literature, the zeta potential at different materials depends highly on the

electrolyte type and the ionic strength. As can be seen in Figure 5.13, the difference in

zeta potential between the PDMS/PDMS channels and the PDMS/glass varies from

buffer to buffer, being higher at PDMS/glass channels in some of the buffers and lower in

others [101]. For example, in Figure 5.14 for N-(2-acetamido)-2-aminoethanesulfonic

acid (ACES) buffer, it is shown that at high ionic strength the zeta potentials at

PDMS/glass and PDMS/PDMS surfaces are comparable but, for low ionic strength, the

zeta potential in PDMS/glass is much higher than the one in PDMS/ PDMS [62].

Figure 5.14: EO mobility as function of ionic strength for ACES buffer using fused

silica, PDMS/glass and PDMS/ PDMS channels [62].

87

Table 5.2 summarizes the zeta potential and EO mobility values used in the simulation.

The TBE and PBS mobility values were extracted from Figure 5.13, using the light color

columns (total length method [101]) and considering that the PDMS/glass value is the

average of PDMS/PDMS and glass/glass values to consequently obtain the zeta potential

at the glass surface. In the case of TE buffer, PDMS/ PDMS data for that buffer was not

provided in the Figure 5.13. Therefore, it was assumed that the zeta potential decreases to

-50 mV on PDMS surfaces and then this value was used to calculate the glass zeta

potential. The PDMS/PDMS mobility obtained for TE buffer was 3.52×10-4

cm2∙V

-1∙s

-1

which agrees with Figure 5.14 for low ionic strengths (TE ionic strength=13 mM).

The EO mobilities showed in table 5.2 were used as boundary conditions for the stokes

flow while the electrical boundary conditions were 0 and +3VDC in the left and right

electrode, respectively, as shown in Figure 5.12, and defining electrical insulation for the

glass and PDMS walls.

Table 5.2. Zeta Potentials and EO mobilities at different surfaces for each buffer

used

Figures 5.15, 5.16 and 5.17 show the simulations results obtained for TBE buffer, TE

buffer and PBS buffer, respectively. The potential and the electric field streamline are

Solution TBE TE PBS

Material

Zeta

Potential

(mV)

EO

Mobility

(cm2/Vs)

Zeta

Potential

(mV)

EO

Mobility

(cm2/Vs)

Zeta

Potential

(mV)

EO

Mobility

(cm2/Vs)

PDMS/Glass -48 3.40×10-4

-68 4.8×10-4

-22 1.55×10-4

PDMS -49 3.45×10-4

-50 3.52×10-4

-23 1.62×10-4

Glass -47 3.3×10-4

-86 6.05×10-4

-21 1.48×10-4

88

shown in Figures 5.15a, 5.16a and 5.17a, where the field density increases near the

electrode edges and being equal in three cases.

Figures 5.15b, 5.16b and 5.17b illustrate the electric field over the cross section area and

the streamlines represent the mean velocity of the fluid. Using different buffers, the

generation of two recirculating flow is observed in all of them, where it can be noticed

that the vortices are less concentric as the difference between the top and bottom

electroosmotic mobilities increases. In addition, these profiles show that near the

electrode edge, where the electric field is higher, the formation of small circulation path

can be seen as part of a secondary flow.

The bottom graphs, Figures 5.15c, 5.16c and 5.17c, represent the velocity field plotted as

a surface plot and velocity arrows showing the velocity profile at different points in the z

direction, along the width of the channel. It can be observed that the velocity at the wall

is negative and has the same direction as the electric field. This flow represents the

movement of the positive charges at the electric double layer near the top and bottom

surfaces as shown in the previous chapter (Figure 4.1). Consequently, the bulk fluid

moves as a result of the drag forces induced by these ions at the wall. However, the

velocity field of the bulk fluid changes direction close to the center of the channel height

allowing the flow recirculation due to the enclosing walls to conserve mass.

The EO flow simulation obtained here is similar to the flow proposed in Figure 4.3a in

Chapter 4. However, the simulation results are more elliptical due to the high aspect ratio

between the width and the height of the device.

89

Figure 5.15: TBE buffer case: (a) potential surface plot and field streamlines, (b) field surface plot and velocity streamlines, (c)

velocity field surface plot and arrow plot.

90

Figure 5.16: TE buffer case: (a) potential surface plot and field streamlines, (b) field surface plot and velocity streamlines, (c)

velocity field surface plot and arrow plot.

91

Figure 5.17: PBS buffer case: (a) potential surface plot and field streamlines, (b) field surface plot and velocity streamlines, (c)

velocity field surface plot and arrow plot.

92

The three cases presented differ from each other by the magnitude and the relative

difference in EO mobility between top and bottom surface which produce a different

velocity magnitude and asymmetry along the channel height.

Figure 5.18: Electroosmotic velocity profiles in z direction (Uz) at the center of the

channel and across the channel height for TBE, TE and PBS buffers.

Among the three cases, the TE buffer generates the largest and the most asymmetric EO

flow as can be observed in Figure 5.18. On the other hand, the velocity generated using

PBS buffer is the lowest one due to the EO mobility reduction in high ionic strength

buffers. Finally, the TBE case is in between the former cases where the EO flow is still

noticeable and approximately symmetric.

To determine the net velocity of DNA traveling within this recirculation flow profile, the

EP velocity of the analyte in the corresponding buffer is added to the EO profile to obtain

the net velocity of the analyte of interest. From Stellwagen et al work [67], the EP

93

mobility of DNA in TBE was found to be 4.5 ×10-4

cm2∙V

-1∙s

-1, while for DNA mobility

in PBS or TE buffer there is no report. However, the DNA mobility in TE is expected to

be just slightly higher that the TBE and much smaller in PBS due to the high salt

concentration. Using the electric field calculated by the COMSOL simulation, the EP

velocity in TBE was found to be 0.62 ×10-3

m∙s-1

which was added to the EO flow to

obtain the net DNA velocity as shown in Figure 5.19. It can be observed there that the net

velocity of the DNA is positive all the points along the height of the channel, as predicted

in Figure 4.3b, and it goes towards the positive electrode where it was captured during

the experiments.

Figure 5.19: Net DNA velocity profile with EO and EP components (UEO+UEP)

across the channel height for TBE buffer.

Moreover, if the EP mobility in TE is considered to be similar to the mobility in TBE,

then the EP velocity would be also approximately 0.62 ×10-3

m∙s-1

. If this EP velocity is

94

added to the EO flow velocity for the TE buffer as it is shown in Figure 5.20, the net

velocity will not always be positive as seen in the TBE case due to the opposing high EO

flow, especially at the glass wall where the EO velocity was found to be higher in

magnitude (-0.83 ×10-3

m∙s-1

) than the estimated EP velocity (~0.62 ×10-3

m∙s-1

). As a

consequence, the net velocity profile have a zero-velocity point near the glass wall where

DNA ultimately migrates to this equilibrium position as predicted in Figure 4.3c.

Figure 5.20: Estimated Net DNA velocity profile with EO and EP components

(UEO+UEP) across the channel height for TE buffer.

Particle Visualization

There is strong evidence that transverse DC fields in combination with axial fields or

pressure driven flows result in a vortical recirculation EO flow profile which follow a

helical path down the length of the channel [79, 80]. In order to observe the recirculating

95

flow, fluorescent uncharged melamine and negatively-charged polystyrene beads

suspended in 1× TE buffer were infused allowing visualization of the recirculation

vortex.

In Figure 5.21, snapshots of a video acquired at the entrance of the device captured the

EO recirculation vortex formation. The charged particles moved quickly to the positive

electrodes (thought to be due to the large particle EP mobility). However, the melamine

particles recirculated with the EO flow profile moving down the channel in a helical

fashion showing the recirculation vortex formed near the center of the channel.

Figure 5.21: Snapshots of a helical flow formation using fluorescent uncharged

melamine and negatively-charged polystyrene beads suspended in TE buffer under

transverse field.

E

Flow

100 m

(a) (b)

(c) (d)

96

The visualization shown in Figure 5.21 demonstrates the ability to trap species inside of

the channel when a transverse field is applied as a result of the helical electroosmotic

flow and the balance with the electrophoresis force.

In summary, a continuous-flow DNA and protein concentration method is demonstrated

here by employing uniform DC or pulsed electric fields applied across the width of a

microchannel. Negatively charged samples are conjunctly infused in a converging Y-inlet

with a receiving buffer of the same conductivity into a main daughter channel where

coplanar electrodes running axially along the length of the channel are used to apply a

transverse electric field. Depending on the buffer selection, the samples could be

concentrated by electrophoretic forces at the positive bias electrode or trapped at an

equilibrium position within the channel by balancing the electrophoretic (EP) force with a

drag force generated by an induced electroosmotic (EO) flow in the opposite direction.

To the author knowledge this is the first report of such trapping and concentration of

DNA and protein samples using transverse DC electric fields. Such a technique could

have advantages over other electrokinetic preconcentration techniques such as FASS, ITP

pH-mediated stacking, and DEP, because it is a simple technique which can be

implemented in a continuous flow environment and does not require a conductivity, pH,

or field gradient.

In the following chapter, the conclusions for the DNA liquid extraction platform, DNA

and protein trapping and concentration will be addressed. In addition, the future work to

be explored as a consequence of the finding shown in this research is detailed next.

97

Chapter 6

Conclusions and Future Directions

The concluding remarks of this original work are presented next, summarizing the results

achieved and the future work towards the development of an integrated lab-on-chip

device for DNA purification via microfluidic phenol extraction as well as the novel DNA

or protein manipulation by transverse electrokinetic balance.

6.1 Microfluidic DNA purification via phenol extraction

Conclusions

In the first part of this dissertation microfluidic platforms have been designed and

fabricated using microfabrication techniques to demonstrate DNA purification via liquid

phase extraction at the microscale using an aqueous phase containing either protein, DNA

or a complex cell lysate and an immiscible receiving organic (phenol) phase.

Initially, a serpentine device was used to investigate protein partitioning between the

aqueous and organic phase, and DNA purification when both BSA protein and DNA

were mixed in the aqueous phase and infused conjunctly with the phenol phase. This two-

phase system was studied using both stratified and droplet-based flow conditions. The

98

droplet based flow resulted in a significant improvement of BSA protein partitioning

(>91% extracted from the aqueous phase into the organic phase) due to the convective

flow recirculation inside each droplet improving material transport to the organic-

aqueous interface.

A second device was designed and fabricated after the preliminary tests with the

serpentine device to specifically extract plasmid DNA from bacterial lysates using only

droplet-based flows. The plasmid recovery using the microdevice was high (>92%) and

comparable to the recovery achieved using commercial DNA purification kits and

standard macro-scale phenol extraction.

These results represent the initial steps towards the miniaturization of an efficient on-chip

DNA sample preparation using phenol extraction which could be integrated with post-

extraction DNA manipulations for integrated genomic analysis modules.

Future Work

For the first time, phenol extraction was realized at the microscale with high efficiency

and comparable with other standard DNA purification methods. In addition, the

purification was performed in a continuous manner without the requirements of valves or

washing and elution steps, which could allow integration with a pre-conditioning module

for cell lysis, and post-extraction manipulation and analysis modules such as enzymatic

restriction enzyme digestion and micro-electrophoretic separation.

However, some issues associated with this method must be solved in order to advance

full integrated on-chip integration. First, following the DNA extraction module, the

99

droplets need to be separated into their individual phases where the phenol phase is sent

to waste while the aqueous phase containing the DNA can be easily removed and used in

downstream processes. For the droplet-based experiments shown here, the two phases

were collected together at the outlet and separated by centrifugation outside of the device.

One approach to separate the two phases at a bifurcating outlet could be based on the

physical properties and favorable energetic interactions of each solvent with a

surrounding wall. By selectively coating one of the outlet channels with a hydrophobic

self assembled monolayer (SAM) [54], the phenol phase can be directed into one outlet

while the organic phase would be directed to another.

A second approach to investigate is the possibility to extract the organic phase by a pillar

array where phenol can be evacuated but the aqueous droplet will not penetrate due to the

high interfacial force [102].

After the phase separation is achieved, the phenol trace in the aqueous phase needs to be

addressed because it would interfere with downstream steps. The second part of this

dissertation was initially proposed to solve this phenol contamination. Next, a summary

of the conclusions and future work of the second part of this work is presented and how

the phenol decontamination can be approached using a buffer exchange module using

transverse electrokinetic forces.

100

6.2 DNA trapping by electrokinetic forces

Conclusions

Sample pre-concentration can be a critical element to improve sensitivity of integrated

microchip assays. In this work a converging Y-inlet microfluidic channel with integrated

coplanar electrodes was used to investigate transverse DNA and protein migration under

uniform direct current (DC) electric fields to assess the ability to concentrate a sample

prior to other enzymatic modifications or capillary electrophoretic separations.

Employing a pressure-driven flow to perfuse the microchannel, negatively charged

samples diluted in low and high ionic strength buffers were co-infused with a receiving

buffer of the same ionic strength into a main daughter channel.

Experimental results demonstrated that, depending of the buffer selection, different DNA

migration and accumulation dynamics were seen. Charged analytes could traverse the

channel width and accumulate at the positive bias electrode in case of a high ionic

strength buffer with low electroosmotic mobility and high electrophoretic mobility, or

migrated towards an equilibrium position within the channel when using a low ionic

strength buffer with high electroosmotic mobility and high electrophoretic mobility. The

various migration behaviors are the result of a balance between the electrophoretic force

and a drag force induced by a recirculating electroosmotic flow generated across the

channel width due to the bounding walls.

Under continuous flow conditions, DNA samples were concentrated up to 4-folds by

trapping in electrodes and 3-folds by balancing transverse electrokinetic forces. The

trapping mechanism was demonstrated and reproduced several times using low voltages

101

(3V) to produce DC fields but also it was observed when using pulsed fields. The DNA

reached the steady state position in less than 15 seconds.

The electrokinetic trapping technique presented here is a simple technique which just

involves a set of coplanar electrodes and low-voltage requirements allowing two different

behaviors depending of the electrolyte solution selection.

Future work

This work demonstrated, for the first time, that DNA and protein samples could be

trapped and concentrated under continuous flow conditions by balancing a DC

electrophoretic force with the electroosmotic flow force. Such a mechanism could be

used as a pre-concentration method prior to DNA loading into a CE separation column.

The trapping behavior makes this technique suitable for concentration enhancements of

dilute solutions to improve detection.

Approaches will be taken to suppress the non-specific absorption of DNA and/or protein

onto the channel surfaces without modifying the wall zeta potential to allow better DNA

concentration quantification. This work will also seek to infuse a low concentration DNA

solution below the fluorescence detection limit in order to concentrate the DNA to a level

that can be detected via epifluorescence microscopy.

A better understanding of the trapping mechanism needs to be addressed to provide the

possibility of controlling concentration enhancement and location across the channel in

order to expand this technique to concentration or separation of other analytes for

downstream processes. This could be explored by different ways. First, the ionic strength

102

of TE buffer can be varied to investigate the DNA trapping as the ionic strength increases

and the electroosmotic drag force gets weaker. Second, the DNA electrophoretic force

can be also modified by varying the NaCl salt concentration to study the force balance.

Third, the development of high fidelity numerical tools can simulate complex, nonlinear

electrokinetic phenomena by coupling electromigration fluxes of all species with the

predicted electroosmotic and hydrodynamic flow profiles.

Additionally, the transverse electrokinetic force can be used to solve the phenol

contamination problem found in the aqueous phase in the first part of this work. As

mentioned previously, high light absorption at low wavelengths (<300 nm) was detected

in aliquots of the aqueous phase following the phenol extraction impeding DNA purity

quantification by spectrophotometry. To eliminate this contamination, a buffer exchange

module can be added after the phase separation to wash and collect DNA in a clean

buffer. The device presented in the second part of this work can be used to force the DNA

sample to travel to the opposite electrode into a clean receiving buffer. For this buffer

exchange application, the device walls needs to be treated first to avoid the generation of

electroosmotic flow and assure that DNA is only affected by the electrophoretic force.

Finally, future work will also explore the utilization of the helical electroosmotic flow to

improve the efficiency of DNA modifying enzyme reactions (restriction endonucleases,

ligases, polymerases, etc.) by infusing them in the receiving buffer and mixing them with

the DNA solution for an integrated reaction control.

103

Bibliography

1. Lui, C., N. Cady, and C. Batt, Nucleic Acid-based Detection of Bacterial

Pathogens Using Integrated Microfluidic Platform Systems. Sensors, 2009. 9(5):

p. 3713-3744.

2. Lagally, E.T., C.A. Emrich, and R.A. Mathies, Fully integrated PCR-capillary

electrophoresis microsystem for DNA analysis. Lab Chip, 2001. 1(2): p. 102-107.

3. Easley, C.J., et al., A fully integrated microfluidic genetic analysis system with

sample-in-answer-out capability. P. Natl. Acad. Sci. USA, 2006. 103(51): p.

19272-19277.

4. Kohlheyer, D., et al., Miniaturizing free-flow electrophoresis - a critical review.

Electrophoresis, 2008. 29(5): p. 977-993.

5. Sommer, G.J. and A.V. Hatch, IEF in microfluidic devices. Electrophoresis, 2009.

30(5): p. 742-757.

6. Sambrook, J., E.F. Fritsch, and T. Maniatis, Molecular Cloning: A Laboratory

Manual. 1989, NY: Cold Spring Harbor Laboratory Press.

7. Qiagen, QIAprep Miniprep Handbook. 2003: Qiagen Distributors

8. Christel, L.A., et al., Rapid, automated nucleic acid probe assays using silicon

microstructures for nucleic acid concentration. J. Biomech Eng, 1999. 121(1): p.

22-7.

9. McMillan, W.A., et al. Application of advanced microfluidics and rapid PCR to

analysis of microbial targets. in Proceedings of the 8th international symposium

on microbial ecology. 1999.

10. Breadmore, M.C., et al., Microchip-Based Purification of DNA from Biological

Samples. Anal. Chem., 2003. 75(8): p. 1880-1886.

11. Hong, J.W., et al., A nanoliter-scale nucleic acid processor with parallel

architecture. Nature Biotechnology, 2004. 22(4): p. 435-9.

12. Wu, Q.R., et al., Microchip-based macroporous silica sol-gel monolith for

efficient isolation of DNA from clinical samples. Anal. Chem., 2006. 78(16): p.

5704-5710.

13. Wen, J., et al., Microfluidic-Based DNA Purification in a Two-Stage, Dual-Phase

Microchip Containing a Reversed-Phase and a Photopolymerized Monolith. Anal.

Chem., 2007. 79(16): p. 6135-6142.

104

14. Kulinski, M.D., et al., Sample preparation module for bacterial lysis and isolation

of DNA from human urine. Biomed. Microdevices, 2009. 11(3): p. 671-678.

15. Bhattacharyya, A. and C.M. Klapperich, Microfluidics-based extraction of viral

RNA from infected mammalian cells for disposable molecular diagnostics.

Sensors Actuat B: Chem, 2008. 129(2): p. 693-698.

16. Chen, X., F. Cui da, and C.C. Liu, On-line cell lysis and DNA extraction on a

microfluidic biochip fabricated by microelectromechanical system technology.

Electrophoresis, 2008. 29(9): p. 1844-1851.

17. Hagan, K.A., et al., Chitosan-Coated Silica as a Solid Phase for RNA Purification

in a Microfluidic Device. Anal. Chem., 2009. 81(13): p. 5249-5256.

18. Witek, M.g.A., et al., Purification and preconcentration of genomic DNA from

whole cell lysates using photoactivated polycarbonate (PPC) microfluidic chips.

Nucleic Acids Res. 34(10): p. e74.

19. Reddy, V. and J.D. Zahn, Interfacial stabilization of organic-aqueous two-phase

microflows for a miniaturized DNA extraction module. J Colloid Interface Sci,

2005. 286(1): p. 158-65.

20. Mao, X., S. Yang, and J. Zahn. Experimental Demonstration and Numerical

Simulation of Organic-Aqueous Liquid Extraction Enhanced by Droplet

Formation in a Microfluidic Channel. in Proceedings of the ASME-IMECE. 2006.

Chicago, IL.

21. Zahn, J. and V. Reddy, Two phase micromixing and analysis using

electrohydrodynamic instabilities. Microfluidics Nanofluidics, 2006. 2(5): p. 399-

415.

22. Morales, M. and J. Zahn. Development of a Diffusion Limited Microfluidic

Module for DNA Purification via Phenol Extraction. in Proceedings of the ASME

IMECE. 2008. Boston, MA

23. Morales, M. and J. Zahn, Droplet enhanced microfluidic-based DNA purification

from bacterial lysates via phenol extraction. Microfluidics Nanofluidics, 2010.

24. Albertsson, P., Partition of cell particles and macromolecules. Third ed. 1986,

New York, NY: Wiley.

25. Shui, L., J.C.T. Eijkel, and A. van den Berg, Multiphase flow in microfluidic

systems - Control and applications of droplets and interfaces. Advances in

Colloid and Interface Science, 2007. 133(1): p. 35-49.

26. Christopher, G.F. and S.L. Anna, Microfluidic methods for generating continuous

droplet streams. J. Phys. D Appl. Phys., 2007. 40(19): p. R319-R336.

27. Shui, L., J.C.T. Eijkel, and A. van den Berg, Multiphase flow in micro- and

nanochannels. Sensors Actuat B: Chem, 2007. 121(1): p. 263-276.

28. Shui, L., et al., Multiphase flow in lab on chip devices: A real tool for the future?

Lab Chip, 2008. 8(7): p. 1010-1014.

105

29. Atencia, J. and D.J. Beebe, Controlled microfluidic interfaces. Nature, 2005.

437(7059): p. 648-655.

30. Helen Song, D.L.C., Rustem F. Ismagilov,, Reactions in Droplets in Microfluidic

Channels. Angewandte Chemie International Edition, 2006. 45(44): p. 7336-

7356.

31. Squires, T.M. and S.R. Quake, Microfluidics: Fluid physics at the nanoliter scale.

Rev. Mod. Phys., 2005. 77(3): p. 977-50.

32. Nguyen, N.T. and S.T. Wereley, Fundamentals And Applications of Microfluidics

Second ed. 2006: Artech House Publishers.

33. Purcell, E.M., Life at low Reynolds number. American Journal of Physics, 1977.

45(1): p. 3-11.

34. Brody, J.P., et al., Biotechnology at low Reynolds numbers. Biophys. J., 1996.

71(6): p. 3430-3441.

35. Dreyfus, R., P. Tabeling, and H. Willaime, Ordered and Disordered Patterns in

Two-Phase Flows in Microchannels. Phys. Rev. Lett., 2003. 90(14): p. 144505.

36. Xu, J., et al., Correlations of droplet formation in T-junction microfluidic devices:

from squeezing to dripping. Microfluidics Nanofluidics, 2008. 5(6): p. 711-717.

37. Guillot, P. and A. Colin, Stability of parallel flows in a microchannel after a T

junction. Phys. Rev. E, 2005. 72(6): p. 066301-4.

38. Tice, J.D., et al., Formation of Droplets and Mixing in Multiphase Microfluidics

at Low Values of the Reynolds and the Capillary Numbers. Langmuir, 2003.

19(22): p. 9127-9133.

39. Tice, J., A. Lyon, and R. Ismagilov, Effects of viscosity on droplet formation and

mixing in microfluidic channels. Analytica Chimica Acta, 2004. 507: p. 73-77.

40. Hibara, A., et al., Integrated Multilayer Flow System on a Microchip. Analytical

Sciences, 2001. 17(1): p. 89-93.

41. Surmeian, M., et al., Three-Layer Flow Membrane System on a Microchip for

Investigation of Molecular Transport. Anal. Chem., 2002. 74(9): p. 2014-2020.

42. Tokeshi, M., et al., Continuous-flow chemical processing on a microchip by

combining microunit operations and a multiphase flow network. Anal. Chem.,

2002. 74(7): p. 1565-1571.

43. Link, D.R., et al., Geometrically Mediated Breakup of Drops in Microfluidic

Devices. Phys. Rev. Lett., 2004. 92(5): p. 054503.

44. Masumi Yamada, V.K., Megumi Nakashima, Jun'ichi Edahiro, Minoru Seki,,

Continuous cell partitioning using an aqueous two-phase flow system in

microfluidic devices. Biotechnology and Bioengineering, 2004. 88(4): p. 489-494.

45. Castell, O.K., C.J. Allender, and D.A. Barrow, Continuous molecular enrichment

in microfluidic systems. Lab Chip, 2008. 8(7): p. 1031-1033.

106

46. Xia, H.M., et al., Chaotic micromixers using two-layer crossing channels to

exhibit fast mixing at low Reynolds numbers. Lab on a Chip, 2005. 5(7): p. 748-

755.

47. Stroock, A.D., et al., Chaotic Mixer for Microchannels. Science, 2002.

295(5555): p. 647-651.

48. Aref, H., The development of chaotic advection. Physics of Fluids, 2002. 14(4): p.

1315-1325.

49. Liu, R.H., et al., Passive mixing in a three-dimensional serpentine microchannel.

Microelectromechanical Systems, Journal of, 2000. 9(2): p. 190-197.

50. Song, H., et al., Experimental test of scaling of mixing by chaotic advection in

droplets moving through microfluidic channels. Applied Physics Letters, 2003.

83(22): p. 4664-4666.

51. Sober, H. and R. Harte, Handbook of Biochemistry. Selected Data for Molecular

Biology. 1st ed. 1968, Cleveland, Ohio: The Chemical Rubber Company,.

52. Xia, Y. and G.M. Whitesides, Soft Lithography. Angew. Chem. Int. Edit., 1998.

37(5): p. 550-575.

53. Haubert, K., T. Drier, and D. Beebe, PDMS bonding by means of a portable, low-

cost corona system. Lab Chip, 2006. 6(12): p. 1548-1549.

54. Hibara, A., et al. Novel Two-Phase Flow Control Concept and Multistep

Extraction Microchip. in Proceeding of the 12th International Conference on

Miniuturized Systems for Chemistry and Life Sciences (TAS 2008). 2008. San

Diego, CA.

55. Song, H., D. Chen, and R. Ismagilov, Reactions in Droplets in Microfluidic

Channels. Angew. Chem. Int. Edit., 2006. 45(44): p. 7336-7356.

56. Wen, J., et al., Purification of nucleic acids in microfluidic devices. Anal. Chem.,

2008. 80(17): p. 6472-6479.

57. Hunter, R.J., Foundations of Colloid Science. Vol. 1. 1987, Oxford: Oxford

University.

58. Probstein, R.F., Physicochemical Hydrodynamics. 2003, Hoboken: Wiley.

59. Tandon, V., et al., Zeta potential and electroosmotic mobility in microfluidic

devices fabricated from hydrophobic polymers: 1. The origins of charge.

Electrophoresis, 2008. 29(5): p. 1092-1101.

60. Kirby, B.J. and E.F. Hasselbrink, Zeta potential of microfluidic substrates: 1.

Theory, experimental techniques, and effects on separations. Electrophoresis,

2004. 25(2): p. 187-202.

61. Tseng, W.L., et al., Effect of ionic strength, pH and polymer concentration on the

separation of DNA fragments in the presence of electroosmotic flow. J.

Chromatogr A, 2000. 894(1-2): p. 219-230.

107

62. Spehar, A.-M., et al., Electrokinetic characterization of poly(dimethylsiloxane)

microchannels. Electrophoresis, 2003. 24(21): p. 3674-3678.

63. Gaudioso, J. and H.G. Craighead, Characterizing electroosmotic flow in

microfluidic devices. J. Chromatogr A, 2002. 971(1-2): p. 249-253.

64. Wu, D., J. Qin, and B. Lin, Electrophoretic separations on microfluidic chips. J.

Chromatogr A, 2008. 1184(1-2): p. 542-559.

65. Wu, D., J. Qin, and B. Lin, Self-assembled epoxy-modified polymer coating on a

poly(dimethylsiloxane) microchip for EOF inhibition and biopolymers separation.

Lab Chip, 2007. 7(11): p. 1490-1496.

66. Xiao, D., T.V. Le, and M.J. Wirth, Surface Modification of the Channels of

Poly(dimethylsiloxane) Microfluidic Chips with Polyacrylamide for Fast

Electrophoretic Separations of Proteins. Anal. Chem., 2004. 76(7): p. 2055-2061.

67. Stellwagen, N., C., C. Gelfi, and P.G. Righetti, The free solution mobility of DNA.

Biopolymers, 1997. 42(6): p. 687-703.

68. Chang, H.C. and L.Y. Yeo, Electrokinetically driven Microfluidics and

Nanofluidics 2010, New York: Cambridge University Press.

69. Viovy, J.L., Electrophoresis of DNA and other polyelectrolytes: Physical

mechanisms. Rev. Mod. Phys., 2000. 72(3): p. 813-872.

70. Stellwagen, E. and N.C. Stellwagen, The free solution mobility of DNA in Tris-

acetate-EDTA buffers of different concentrations, with and without added NaCl.

Electrophoresis, 2002. 23(12): p. 1935-1941.

71. Stellwagen, E. and N.C. Stellwagen, Probing the Electrostatic Shielding of DNA

with Capillary Electrophoresis. Biophys. J., 2003. 84(3): p. 1855-1866.

72. Chang, C.C. and R.J. Yang, Electrokinetic mixing in microfluidic systems.

Microfluidics Nanofluidics, 2007. 3(5): p. 501-525.

73. Qian, S. and H.H. Bau, A Chaotic Electroosmotic Stirrer. Anal. Chem., 2002.

74(15): p. 3616-3625.

74. Stroock, A.D., et al., Patterning Electro-osmotic Flow with Patterned Surface

Charge. Phys. Rev. Lett., 2000. 84(15): p. 3314.

75. Glasgow, I., J. Batton, and N. Aubry, Electroosmotic mixing in microchannels.

Lab Chip, 2004. 4(6): p. 558-562.

76. Song, H., et al., Chaotic mixing in microchannels via low frequency switching

transverse electroosmotic flow generated on integrated microelectrodes. Lab

Chip, 2010. 10(6): p. 734-740.

77. Fu, L.-M., et al., A novel microfluidic mixer utilizing electrokinetic driving forces

under low switching frequency. Electrophoresis, 2005. 26(9): p. 1814-1824.

78. Sasaki, N., T. Kitamori, and H.-B. Kim, AC electroosmotic micromixer for

chemical processing in a microchannel. Lab Chip, 2006. 6(4): p. 550-554.

108

79. Lynn, N.S., C.S. Henry, and D.S. Dandy, Microfluidic mixing via transverse

electrokinetic effects in a planar microchannel. Microfluidics Nanofluidics, 2008.

5(4): p. 493-505.

80. Kim, S.J., I.S. Kang, and B.J. Yoon, Electroosmotic helical flow produced by

combined use of longitudinal and transversal electric fields in a rectangular

microchannel. Chem. Eng. Commun., 2006. 193(9): p. 1075-1089.

81. Burgi, D.S. and R.L. Chien, Optimization in sample stacking for high-

performance capillary electrophoresis. Anal. Chem., 1991. 63(18): p. 2042-2047.

82. Bharadwaj, R. and J.G. Santiago, Dynamics of field-amplified sample stacking. J.

Fluid Mech, 2005. 543(-1): p. 57-92.

83. Gong, M., et al., On-Line Sample Preconcentration Using Field-amplified

Stacking Injection in Microchip Capillary Electrophoresis. Anal. Chem., 2006.

78(11): p. 3730-3737.

84. Jung, B., R. Bharadwaj, and J.G. Santiago, Thousandfold signal increase using

field-amplified sample stacking for on-chip electrophoresis. Electrophoresis,

2003. 24(19-20): p. 3476-3483.

85. Huang, H., et al., On-line isotachophoretic preconcentration and gel

electrophoretic separation of sodium dodecyl sulfate-proteins on a microchip.

Electrophoresis 2005. 26(11): p. 2254-2260.

86. Wainright, A., et al., Sample pre-concentration by isotachophoresis in

microfluidic devices. J. Chromatogr A, 2002. 979(1-2): p. 69-80.

87. Breadmore, M.C. and J.P. Quirino, 100 000-Fold Concentration of Anions in

Capillary Zone Electrophoresis Using Electroosmotic Flow Controlled

Counterflow Isotachophoretic Stacking under Field Amplified Conditions. Anal.

Chem., 2008. 80(16): p. 6373-6381.

88. Schonfeld, F., et al., Transition zone dynamics in combined isotachophoretic and

electro-osmotic transport. Physics of Fluids, 2009. 21(9): p. 11.

89. Monton, M.R.N., et al., Dynamic pH junction technique for on-line

preconcentration of peptides in capillary electrophoresis. J. Chromatogr A, 2005.

1079(1-2): p. 266-273.

90. Kohlheyer, D., et al., Free-flow zone electrophoresis and isoelectric focusing

using a microfabricated glass device with ion permeable membranes. Lab Chip,

2006. 6(3): p. 374-380.

91. Osafune, T., H. Nagata, and Y. Baba, Influence of the pH on Separating DNA by

High-Speed Microchip Electrophoresis. Anal. Sci., 2004. 20(6): p. 971-974.

92. Asbury, C.L., A.H. Diercks, and G. van den Engh, Trapping of DNA by

dielectrophoresis. Electrophoresis, 2002. 23(16): p. 2658-2666.

93. Asbury, C.L. and G. van den Engh, Trapping of DNA in nonuniform oscillating

electric fields. Biophys J., 1998. 74(2): p. 1024-1030.

109

94. Wang, L., et al., Dielectrophoresis switching with vertical sidewall electrodes for

microfluidic flow cytometry. Lab Chip, 2007. 7(9): p. 1114-1120.

95. Fiedler, S., et al., Dielectrophoretic Sorting of Particles and Cells in a

Microsystem. Anal. Chem., 1998. 70(9): p. 1909-1915.

96. Du, J.-R., et al., Long-range and superfast trapping of DNA molecules in an ac

electrokinetic funnel. Biomicrofluidics, 2008. 2(4): p. 044103-10.

97. Liu, P., et al., Integrated DNA purification, PCR, sample cleanup, and capillary

electrophoresis microchip for forensic human identification. Lab Chip. 11(6): p.

1041-1048.

98. Hatch, A.V., et al., Integrated Preconcentration SDS-PAGE of Proteins in

Microchips Using Photopatterned Cross-Linked Polyacrylamide Gels. Anal.

Chem., 2006. 78(14): p. 4976-4984.

99. Foote, R.S., et al., Preconcentration of Proteins on Microfluidic Devices Using

Porous Silica Membranes. Anal. Chem., 2004. 77(1): p. 57-63.

100. Butler, J.N., Ionic Equilibrium: Solubility and PH Calculations. 1998 New York:

Wiley.

101. Almutairi, Z.A., et al., A Y-channel design for improving zeta potential and

surface conductivity measurements using the current monitoring method.

Microfluidics Nanofluidics, 2009. 6(2): p. 241-251.

102. Niu, X.Z., et al., Droplet-based compartmentalization of chemically separated

components in two-dimensional separations. Chemical Communications,

2009(41): p. 6159-6161.

103. Shrewsbury, P.J., S.J. Muller, and D. Liepmann, Effect of Flow on Complex

Biological Macromolecules in Microfluidic Devices. Biomed. Microdevices,

2001. 3(3): p. 225-238.

110

Appendix A

Microfluidic Channel and Electrode Fabrication

Photolithography process using SU-8 2010 and 2050 Negative

Photoresist

Soak substrate (glass or silicon) in acetone, isopropanol and deionized (DI) water

baths for 10 minutes each.

Rinse with DI water and blow dry with nitrogen or filtered air.

Dehydrated the glass slides at a 150°C for 30 minutes in an oven and let to cool

before spinning.

Dispense the photoresist on the substrate trying to avoid bubbles and overload the

substrate that can lead to chuck groove obstruction.

Spin conditions: for 20 μm (SU-8 2010): 30 sec at 1000 rpm, for 50 μm (SU-8

2050): 30 sec at 1600 rpm.

Soft-bake for SU-8 2010: 1 minute at 65°C and 4 minutes at 95°C on a hotplate.

For SU-8 2050: 2 minutes at 65°C and 6 minutes at 95°C.

111

Expose photoresist to UV light using the appropriate recipe for the substrate

selected.

Post-Exposure bake for SU-8 2010: 1 minute at 65°C and 2 minutes at 95°C on a

hotplate. For SU-8 2050: 1 minute at 65°C and 4 minutes at 95°C on a hotplate.

Develop photoresist by immersing slide in SU-8 developer with gentle agitation.

For SU-8 2010: 3 minutes. SU-8 2050: 8 minutes.

Gently rinse the slide with isopropanol.

Gently dry the glass slide using an air gun.

Silanization

This process is convenient to help the release of the PDMS from the master and

preserve the photoresist.

Place the clean master in a vacuum chamber together with a clean glass slide with

100 μl of trichloro (1, 1, 2, 2-perfluoocytl) silane for 30 minutes.

Soft lithography process

Mix 10 parts of PDMS base and 1 part of PDMS crosslinker.

Pour PDMS mix on top of the master.

Place it in vacuum chamber to remove air bubbles.

Cure for 1 hour at 65 C.

112

Punch hole on all inlets and outlets.

Clean and rinse PDMS and glass surfaces with acetone, then isopropanol, then DI

water.

Dry both part thoroughly with filtered compressed air.

Bond by treating bonding surfaces with corona discharge or using a plasma

generator at for 1 minute.

Align PDMS channel over glass slide. Place the device in oven at 65°C for 1 hour

to allow bonding.

Photolithography process using S1818 (Shipley)

Clean the glass substrate as in the SU-8 protocol.

Deposit 3-4 drops of hexamethyldisilazane (HDMS) to prime the surface and spin

at 3000 rpm.

Deposit S1818 and spin at 3000 rpm.

Soft bake at 120°C for 4 minutes and let it cool.

Expose the photoresist using the Shipley recipe (applying 150 mJ/cm2).

Develop in MF-319 for 1-4 minutes.

Rinse with DI water and blow dry.

Hard bake: place the substrate in an oven at 120°C for 30 minutes. Let it cool.

113

Glass etching: rinse the slide in a bath of buffered hydrofluoric acid (5:1) for 1

minute (Caution: special safety standard operating procedure is required to handle

hydrofluoric acid)

Metal Deposition

Turn on the chiller, open air valve and vent.

Load the samples.

Start to pump down and allow pressures to reach between 1x10-6

to 5x10-7

torr.

Then set turbo pump speed to 50% and let it spin down to 80% or less before

opening up the argon valve.

Start platen rotation (10RPM). Turn on motor and select FWD.

Set the appropriate gas flow rates on the computer. The mass flow controller for

argon is MFC1 and set it to 100 sccm. Then open up the valve on the screen.

Once the turbo pump reaches 50% and the pressure has stabilized, set deposition

power settings for the sputtering gun to be used. Turn on SWITCH for the

appropriate gun, turn on POWER, then wait 30 sec to 1 min to remove impurities

from the target surface.

Then open the SHUTTER while starting the stopwatch. Sputtering settings:

titanium: 4.5 minutes at 200 W, platinum: 10 minutes at 200 W.

When ready, stop deposition: in this order, close the SHUTTER, turn off the

POWER, then turn off the gun SWITCH (all done on-screen).

114

Close the gas switch on-screen and shut off argon valve on the wall, behind the

machine (leave the air valve ON). Stop all-motion button.

Wait 5 minutes to let the gas get out of the chamber and then vent.

Remove platen from machine and remove specimens from the platen. Place platen

back into the machine.

Start pumping and let complete the pumping cycle.

Turn off these components in this SPECIFIC order: Turbo pump, valve, roughing

pump.

Shut all of the gas valves on the wall, including the house air and turn off

recirculation pump.

115

Appendix B

Sample Preparation and Measurements

BSA Labeling

First, 5-6-carboxy-tetramethylrhodamine (TAMRA) succinimidyl ester dye (Fluka,

Switzerland) is dissolved in Dimethyl Sulfoxide (DMSO) (Amresco, Solon, OH) to a

concentration of 10 mg/ml. BSA is separately dissolved in PBS to a concentration of 50

mg/ml. Then, the amount of dye to be added to the BSA solution is calculated using the

following equation:

[ ] 100BSA BSARhod Rhod

BSA

C VV l MW MR

MW

(B.1)

where CBSA is the concentration of BSA in mg/ml and VBSA is the volume in ml of the

BSA solution. MWBSA is the molecular weight of the protein, 66000 Daltons, and

MWRhod is the molecular weight of the dye, 527.53 Daltons. MR is the molar ratio of dye

to protein. A ratio of 10 was used for this labeling.

The dye solution is added to the protein solution and covered with aluminum foil to keep

it out of light. The solution is mixed using a rotary shaker for 30 minutes.

116

Then, the labeled solution needs to be filtered through a PD-10 desalting column

(Sephadex G-25M; GE Healthcare) to remove the free dye from the solution. After

removing the top cap of the column and pour off the excess liquid, equilibration of the

column with 25 ml PBS buffer is required. Labeled solution is added to the column, and

then, same amount of buffer is added and the output is discarded. Finally, buffer is eluted

and the output flow is collected with the labeled proteins. The solution could be stored at

-20˚C or diluted to the needed concentration. In this project, the concentration used was

0.3 mg/ml.

Spectrophotometry measurements

The samples were loading into the 1ml spectrophotometer cuvette and the reading was

taking as a scanning from 250 to 600 nm. Then, the protein concentration can be

calculated by the following equation:

280 585[ ( )]BSA BSA

A A K DFC MW

(B.2)

Where A280 and A585 are the UV-light absorption of the protein solution at 280 and 585

nm; DF is the dilution factor; ε is the extinction coefficient that for BSA is approximately

43,824 M-1

cm-1

.

DNA Staining

YOYO-1 was intercalated within the DNA backbone at a ratio of one molecule of dye per

five base pairs of DNA (1:5), similarly as described in Shrewsbury et al paper [103]. The

117

average molecular weight of the base pair used was 650 grams/bp mole. The DNA was

diluted in TE buffer (10 mM tris-HCl, 1 mM EDTA, from Rockland Inc., Gilbertsville,

PA). The dye solution was added conjunctly with β-mercaptoethanol, an oxygen

scavenger that protect DNA-dye complex from oxygen radicals, at a final concentration

of 4% (v/v). The incubation of DNA and the dye occurred at room temperature in the

dark and for a minimum of two hours. The final DNA concentration was 0.075 μg/μl.

Bacteria Transformation, Bacteria Plate and Liquid Culture

The bacteria transformation was performed using Bio-Rad pGLO bacterial

transformation kit following protocol from Bio-Rad (http://www.bio-

rad.com/LifeScience/pdf/Bulletin_9563.pdf).

The bacteria culture in plates was done in 1.5% (LB) agar, 2% arabinose, 100

µg/ml ampicillin pre-poured plates from Teknova.

Prepare a starter culture by picking a colony from a plate and incubate in tube

with 3ml of Luria Broth (LB), 300 μl of Arabinose at 2% and 3 μl of ampicillin at

100 µg/µl. Incubate for 8 hours.

To make a fresh plate culture, take 50 μl from the starter culture and diluted in

200 μl of LB. Then using a loop, streak the plate as it is shown Qiagen Bench

Guide

(http://www.qiagen.com/literature/benchguide/pdf/1017778_benchguide_chap_1.

pdf). Incubate for 12-14 hours until colonies develop.

118

To make a liquid culture, take 10 μl from the starter culture and diluted in culture

tube with 3ml of LB, 300 μl of Arabinose at 2% and 3 μl of ampicillin at

100µg/µl concentration. Incubate for 12-14 hours.

DNA Gel Electrophoresis

Prepare 0.8 liter of 1× Tris acetate EDTA (TAE) buffer in a cylinder.

In flask, prepare 1% agarose solution by mixing 1 gram and 100 ml of 1× TAE

buffer.

Then heat the solution in a microwave oven until completely melted.

After cooling the solution to about 60˚C, it is poured into a casting tray. Add the

sample comb and allowed to solidify at room temperature.

After the gel has solidified, remove the side walls of the casting tray and place the

casting tray with the gel in the electrophoresis box. Add the rest of the TAE buffer

to cover the gel.

Remove the comb, using care not to rip the bottom of the wells.

Samples containing DNA mixed with loading buffer are then pipetted into the

sample well on the negative electrode side (black electrode).

The lid and power leads are placed on the apparatus, and voltage is applied (120

V). DNA will migrate towards the positive electrode which is usually colored red.

Stop the migration once the loading dye migrated half of the gel length.

Bring the gel to the ethidium bromide bath and turn on the shaker for 15 minutes.

119

Place the gel into the UV illuminator and obtain pictures of the gel.

120

Appendix C

COMSOL Simulation Parameters

Model Tree and Mesh

Select MEMS module, then Microfluidics and Electroosmotic Flow and finally

Stokes Flow and Steady State analysis.

It will open the following branches in the model tree: Stokes Flow and

Conductive Media DC.

Import AutoCAD drawing and mesh it completely and re-mesh for a finer mesh

near the electrode corners.

Subdomain Settings for Stokes Flow

In Physics tab: ρ=1000, η=.001, Fx=Fy=0, Thickness=1.

In Microfluidic tab: εr=80, λ=1e-6.

Boundary Settings for Stokes Flow

Set all boundaries conditions to “Electroosmotic Velocity”.

121

In Electroosmotic Velocity tab: in all set Ex=Ex_emdc2, Ey=Ey_emdc2, electroosmotic

mobility at the electrode equal to zero, electroosmotic mobility at glass and

PDMS walls according to buffer selected (Table 5.2).

Subdomain Settings for Conductive Media DC

In Physics tab: σ isotropic=.0739, thickness=1.

Boundary Settings for Stokes Flow

Select “Electric Insulation” except for the electrodes and set them to 0 V and 3 V

potentials.

Solver

First get the initial value from “Solve” tab. Then go to “Solver parameter” and set

Spooles.

Go to Solver manager, “Initial Value” tab. Click in “Values of variables not

solved for and linearization point” the option “Use setting from initial value

frame”. Then go to “Solve for” tab and select Stokes and Conductive Media, then

apply.

Go to “Script” tab and add current solver settings. Click “Solve using a script”

and solve. This will solve Stokes and Conductive Media DC together.

Repeat these steps in case you want to solve again.

122

Curriculum Vitae

MERCEDES C. MORALES

Education

2012 PhD in Biomedical Engineering, Rutgers, The State University of New

Jersey.

2008 Master of Science in Biomedical Engineering, Rutgers, The State

University of New Jersey.

2004 Electromechanical Engineering degree, concentration in Industrial

Automation, Universidad Nacional de La Pampa (Argentina).

Experience

2010-2011 Teaching Graduate Assistant, Rutgers, The State University of New

Jersey.

2007-2010 Research Graduate Assistant, Rutgers, The State University of New

Jersey.

2004-2005 Research Assistant, Universidad Nacional de La Pampa (Argentina).

Journal Publications

M. C. Morales, H. Lin and J. D. Zahn. “Continuous microfluidic DNA and

protein trapping and concentration by balancing transverse electrokinetic forces”,

Lab on a Chip, 2011, DOI: 10.1039/C1LC20605B.

M. C. Morales, J. D. Zahn. “Droplet Enhanced Microfluidic-based DNA

Purification from Bacterial Lysates via Phenol Extraction”, Microfluidics and

Nanofluidics, 2010, 9:1041-1049.